OneJev 1.6 0.8B Decision Model Results

OneJev 1.6 0.8B

Explore OneJev 1.6 0.8B: exact decision configuration, published specifications, source-linked benchmarks and API tariffs when available.

Published specifications

Provider: OmniJev
Version: 1.6
System type: Trained checkpoint
Access: Open weights
License: Apache-2.0
Inputs: text, image, video
Decision types: Choice, Yes / no, Score
Published size: 0.85B
Active parameters: Not published
Context / input limit: 262,144 configured positions; evaluation and serving limits vary
Announced: 2026-10-17

Hugging Face · Announcement · Website

Model notes

Returns typed probabilities, not generated prose. Published total tensor count is from this exact official HF API; nominal family names can differ. No active-parameter count is inferred. Configured position limit is not a validated long-context capability. Vendor announcement is dated 17 October; official repository uploads began 16 October UTC. Full fine-tune on about 440,000 decision questions, including public training splits. Text, screenshots, images and video use the published PyTorch path; GGUF runtime supports text/images only and its example still points to an older 1.5 checkpoint, which is not this entry. OneJev 1.6 public-index figures are vendor-run with the community kit, not evaluator-private/full results. Community S1MB parameter-count conventions differ from official tensor totals and are not MoE active counts.

Checkpoint revision: 0acd65660ac22ca42f40b8b6a6da3a039590b412

← All decision models · Compare this decision system →

228 recorded observations. Published results retain their source and setup; no Examenos evaluation was run. Missing results are not zero. How to read these comparisons →

Official / vendor · OneJev (vendor measured)

Evaluator-published · S1MB (community measured runs)

OneJev 1.6 0.8B · Evaluator-published · S1MB (community measured runs)
Evaluation	Result	Evidence
S1MB · aegis2-noul-test-v1 · binary_brier	0.2107 Brier	Setup & source S1MB datasets/aegis2 / test · binary_brier. Population 100 cases / 148 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.2107263177229865,"accuracy":0.6756756756756757,"positive_recall":0.4084507042253521,"specificity":0.922077922077922,"precision":0.8285714285714286,"balanced_accuracy":0.6652643131516371,"f1":0.5471698113207547,"true_positive":29,"false_positive":6,"false_negative":42,"true_negative":71,"positive_prevalence":0.4797297297297297,"majority_accuracy_baseline":0.5202702702702703,"constant_brier_baseline":0.24958911614317017,"brier_skill":0.1557071038221517,"baseline_adjusted_skill":0.3305286263032743,"baseline_adjusted_score":0.3305286263032743}. Counts: {"cases": 100, "expected": 148, "succeeded": 148, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · aqua-rat-choice-test-v1 · target_mass_at_prediction	29.00%	Setup & source S1MB datasets/aqua_rat / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.29,"uniform_accuracy_baseline":0.2005,"fixed_answer_accuracy_baseline":0.29,"baseline_adjusted_skill":0,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · arc-choice-test-v1 · target_mass_at_prediction	77.00%	Setup & source S1MB datasets/arc / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.77,"uniform_accuracy_baseline":0.2533333333333333,"fixed_answer_accuracy_baseline":0.32,"baseline_adjusted_skill":0.661764705882353,"baseline_adjusted_score":0.661764705882353}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · arct-choice-test-v1 · target_mass_at_prediction	71.00%	Setup & source S1MB datasets/arct / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.71,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.53,"baseline_adjusted_skill":0.3829787234042552,"baseline_adjusted_score":0.3829787234042552}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · argument-quality-choice-test-v1 · target_mass_at_prediction	66.00%	Setup & source S1MB datasets/argument_quality / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.66,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.51,"baseline_adjusted_skill":0.3061224489795919,"baseline_adjusted_score":0.3061224489795919}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · argument-quality-noul-test-v1 · binary_brier	0.1572 Brier	Setup & source S1MB datasets/argument_quality / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.1572202488982909,"accuracy":0.52617627734,"positive_recall":0.48878006391935774,"specificity":0.5861746423618762,"precision":0.6545763134782608,"balanced_accuracy":0.537477353140617,"f1":0.5596572663900694,"true_positive":30.11051042,"false_positive":15.88948958,"false_negative":31.492882686,"true_negative":22.507117314000002,"positive_prevalence":0.61603393106,"majority_accuracy_baseline":0.61603393106,"constant_brier_baseline":0.07908707655202402,"brier_skill":-0.9879385577600703,"baseline_adjusted_skill":0.07495470628123391,"baseline_adjusted_score":0.07495470628123391}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · babi-nli-noul-test-v1 · binary_brier	0.2079 Brier	Setup & source S1MB datasets/babi_nli / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.20788997939297998,"accuracy":0.66,"positive_recall":0.54,"specificity":0.78,"precision":0.7105263157894737,"balanced_accuracy":0.66,"f1":0.6136363636363636,"true_positive":27,"false_positive":11,"false_negative":23,"true_negative":39,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.16844008242808006,"baseline_adjusted_skill":0.32000000000000006,"baseline_adjusted_score":0.32000000000000006}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · banking77-choice-test-v1 · target_mass_at_prediction	84.00%	Setup & source S1MB datasets/banking77 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.84,"uniform_accuracy_baseline":0.012987012987012986,"fixed_answer_accuracy_baseline":0.05,"baseline_adjusted_skill":0.831578947368421,"baseline_adjusted_score":0.831578947368421}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · bbq-choice-test-v1 · target_mass_at_prediction	78.00%	Setup & source S1MB datasets/bbq / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.78,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":0.6333333333333334,"baseline_adjusted_score":0.6333333333333334}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · boardgameqa-choice-test-v1 · target_mass_at_prediction	39.00%	Setup & source S1MB datasets/boardgameqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.39,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.36,"baseline_adjusted_skill":0.04687500000000004,"baseline_adjusted_score":0.04687500000000004}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · canttalk-noul-test-v1 · binary_brier	0.1271 Brier	Setup & source S1MB datasets/canttalk / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.12710106875919,"accuracy":0.85,"positive_recall":0.78,"specificity":0.92,"precision":0.9069767441860465,"balanced_accuracy":0.8500000000000001,"f1":0.8387096774193549,"true_positive":39,"false_positive":4,"false_negative":11,"true_negative":46,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.49159572496324,"baseline_adjusted_skill":0.7000000000000002,"baseline_adjusted_score":0.7000000000000002}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · civil-comments-noul-test-v1 · binary_brier	0.0944 Brier	Setup & source S1MB datasets/civil_comments / test · binary_brier. Population 100 cases / 700 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.09438058180981233,"accuracy":0.6532202349223994,"positive_recall":0.4173849419213335,"specificity":0.763033805310778,"precision":0.4505971862880278,"balanced_accuracy":0.5902093736160557,"f1":0.4333556517269232,"true_positive":92.82302037533373,"false_positive":113.17697962466627,"false_negative":129.5688559296541,"true_negative":364.4311440703459,"positive_prevalence":0.3177026804356969,"majority_accuracy_baseline":0.6822973195643032,"constant_brier_baseline":0.08872692397768642,"brier_skill":-0.06371975471106972,"baseline_adjusted_skill":0.1804187472321115,"baseline_adjusted_score":0.1804187472321115}. Counts: {"cases": 100, "expected": 700, "succeeded": 700, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · cladder-noul-test-v1 · binary_brier	0.1410 Brier	Setup & source S1MB datasets/cladder / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.14096705868261,"accuracy":0.77,"positive_recall":0.74,"specificity":0.8,"precision":0.7872340425531915,"balanced_accuracy":0.77,"f1":0.7628865979381443,"true_positive":37,"false_positive":10,"false_negative":13,"true_negative":40,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.43613176526956,"baseline_adjusted_skill":0.54,"baseline_adjusted_score":0.54}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · clinc-choice-test-v1 · target_mass_at_prediction	90.73%	Setup & source S1MB datasets/clinc / test · target_mass_at_prediction. Population 151 cases / 151 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.9072847682119205,"uniform_accuracy_baseline":0.006622516556291391,"fixed_answer_accuracy_baseline":0.039735099337748346,"baseline_adjusted_skill":0.9034482758620689,"baseline_adjusted_score":0.9034482758620689}. Counts: {"cases": 151, "expected": 151, "succeeded": 151, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 151 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · commonsensemoralja-noul · binary_brier	0.1883 Brier	Setup & source S1MB datasets/commonsensemoralja-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.18826294545056665,"accuracy":0.7366666666666667,"positive_recall":0.8066666666666666,"specificity":0.6666666666666666,"precision":0.7076023391812866,"balanced_accuracy":0.7366666666666666,"f1":0.7538940809968847,"true_positive":121,"false_positive":50,"false_negative":29,"true_negative":100,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.24694821819773338,"baseline_adjusted_skill":0.47333333333333316,"baseline_adjusted_score":0.47333333333333316}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · contract-nli-choice-test-v1 · target_mass_at_prediction	86.29%	Setup & source S1MB datasets/contract_nli / test · target_mass_at_prediction. Population 100 cases / 1700 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8629411764705882,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.4576470588235294,"baseline_adjusted_skill":0.747288503253796,"baseline_adjusted_score":0.747288503253796}. Counts: {"cases": 100, "expected": 1700, "succeeded": 1700, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · corr2cause-choice-test-v1 · target_mass_at_prediction	27.00%	Setup & source S1MB datasets/corr2cause / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.27,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":-0.21666666666666667,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · creak-noul-test-v1 · binary_brier	0.1971 Brier	Setup & source S1MB datasets/creak / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.19712369239441,"accuracy":0.7,"positive_recall":0.86,"specificity":0.54,"precision":0.6515151515151515,"balanced_accuracy":0.7,"f1":0.7413793103448276,"true_positive":43,"false_positive":23,"false_negative":7,"true_negative":27,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.21150523042236002,"baseline_adjusted_skill":0.3999999999999999,"baseline_adjusted_score":0.3999999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · crows-pairs-choice-test-v1 · target_mass_at_prediction	60.00%	Setup & source S1MB datasets/crows_pairs / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.6,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.52,"baseline_adjusted_skill":0.1666666666666666,"baseline_adjusted_score":0.1666666666666666}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · cvbench-depth-choice · target_mass_at_prediction	78.33%	Setup & source S1MB datasets/cvbench-depth-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.7833333333333333,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5,"baseline_adjusted_skill":0.5666666666666667,"baseline_adjusted_score":0.5666666666666667}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · cvbench-distance-choice · target_mass_at_prediction	59.33%	Setup & source S1MB datasets/cvbench-distance-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.5933333333333334,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.52,"baseline_adjusted_skill":0.15277777777777785,"baseline_adjusted_score":0.15277777777777785}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · cvbench-relation-choice · target_mass_at_prediction	80.67%	Setup & source S1MB datasets/cvbench-relation-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8066666666666666,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5266666666666666,"baseline_adjusted_skill":0.5915492957746479,"baseline_adjusted_score":0.5915492957746479}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · dbpedia-choice-test-v1 · target_mass_at_prediction	90.00%	Setup & source S1MB datasets/dbpedia / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.9,"uniform_accuracy_baseline":0.07142857142857142,"fixed_answer_accuracy_baseline":0.11,"baseline_adjusted_skill":0.8876404494382023,"baseline_adjusted_score":0.8876404494382023}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · defeasible-nli-choice-test-v1 · target_mass_at_prediction	62.00%	Setup & source S1MB datasets/defeasible_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.62,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":0.1739130434782608,"baseline_adjusted_score":0.1739130434782608}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · esci-choice-test-v1 · target_mass_at_prediction	37.00%	Setup & source S1MB datasets/esci / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.37,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.27,"baseline_adjusted_skill":0.13698630136986298,"baseline_adjusted_score":0.13698630136986298}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · ethics-choice-test-v1 · target_mass_at_prediction	66.67%	Setup & source S1MB datasets/ethics / test · target_mass_at_prediction. Population 18 cases / 18 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.6666666666666666,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.6666666666666666,"baseline_adjusted_skill":0,"baseline_adjusted_score":0}. Counts: {"cases": 18, "expected": 18, "succeeded": 18, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 18 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · ethics-noul-test-v1 · binary_brier	0.2427 Brier	Setup & source S1MB datasets/ethics / test · binary_brier. Population 82 cases / 82 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.2427153841711585,"accuracy":0.5975609756097561,"positive_recall":0.6,"specificity":0.5952380952380952,"precision":0.5853658536585366,"balanced_accuracy":0.5976190476190476,"f1":0.5925925925925926,"true_positive":24,"false_positive":17,"false_negative":16,"true_negative":25,"positive_prevalence":0.4878048780487805,"majority_accuracy_baseline":0.5121951219512195,"constant_brier_baseline":0.2498512790005949,"brier_skill":0.028560569543529923,"baseline_adjusted_skill":0.1952380952380952,"baseline_adjusted_score":0.1952380952380952}. Counts: {"cases": 82, "expected": 82, "succeeded": 82, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 82 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · ethos-noul-test-v1 · binary_brier	0.1659 Brier	Setup & source S1MB datasets/ethos / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16587614611253,"accuracy":0.74,"positive_recall":0.76,"specificity":0.72,"precision":0.7307692307692307,"balanced_accuracy":0.74,"f1":0.7450980392156863,"true_positive":38,"false_positive":14,"false_negative":12,"true_negative":36,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.33649541554988005,"baseline_adjusted_skill":0.48,"baseline_adjusted_score":0.48}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · few-nerd-choice-test-v1 · target_mass_at_prediction	63.41%	Setup & source S1MB datasets/few_nerd / test · target_mass_at_prediction. Population 100 cases / 358 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.6340782122905028,"uniform_accuracy_baseline":0.125,"fixed_answer_accuracy_baseline":0.2206703910614525,"baseline_adjusted_skill":0.5304659498207885,"baseline_adjusted_score":0.5304659498207885}. Counts: {"cases": 100, "expected": 358, "succeeded": 358, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · fol-nli-choice-test-v1 · target_mass_at_prediction	41.00%	Setup & source S1MB datasets/fol_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.41,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.39,"baseline_adjusted_skill":0.03278688524590158,"baseline_adjusted_score":0.03278688524590158}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · followir-core17-noul-test-v1 · binary_brier	0.1639 Brier	Setup & source S1MB datasets/followir_core17 / test · binary_brier. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16385931233577503,"accuracy":0.73,"positive_recall":0.3409090909090909,"specificity":0.8397435897435898,"precision":0.375,"balanced_accuracy":0.5903263403263403,"f1":0.35714285714285715,"true_positive":15,"false_positive":25,"false_negative":29,"true_negative":131,"positive_prevalence":0.22,"majority_accuracy_baseline":0.78,"constant_brier_baseline":0.1716,"brier_skill":0.04510890247217347,"baseline_adjusted_skill":0.1806526806526807,"baseline_adjusted_score":0.1806526806526807}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · followir-news21-noul-test-v1 · binary_brier	0.1861 Brier	Setup & source S1MB datasets/followir_news21 / test · binary_brier. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.186128175414395,"accuracy":0.745,"positive_recall":0.08,"specificity":0.9666666666666667,"precision":0.4444444444444444,"balanced_accuracy":0.5233333333333333,"f1":0.13559322033898305,"true_positive":4,"false_positive":5,"false_negative":46,"true_negative":145,"positive_prevalence":0.25,"majority_accuracy_baseline":0.75,"constant_brier_baseline":0.1875,"brier_skill":0.0073163977898933474,"baseline_adjusted_skill":0.046666666666666634,"baseline_adjusted_score":0.046666666666666634}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · followir-robust04-noul-test-v1 · binary_brier	0.0573 Brier	Setup & source S1MB datasets/followir_robust04 / test · binary_brier. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.05732843650944,"accuracy":0.93,"positive_recall":0.47368421052631576,"specificity":0.9779005524861878,"precision":0.6923076923076923,"balanced_accuracy":0.7257923815062518,"f1":0.5625,"true_positive":9,"false_positive":4,"false_negative":10,"true_negative":177,"positive_prevalence":0.095,"majority_accuracy_baseline":0.905,"constant_brier_baseline":0.085975,"brier_skill":0.333196434900378,"baseline_adjusted_skill":0.4515847630125036,"baseline_adjusted_score":0.4515847630125036}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-diverse-choice · target_mass_at_prediction	93.33%	Setup & source S1MB datasets/generalization-japanese-diverse-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.9333333333333333,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.9,"baseline_adjusted_score":0.9}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-diverse-noul · binary_brier	0.0824 Brier	Setup & source S1MB datasets/generalization-japanese-diverse-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.08243896957920001,"accuracy":0.8966666666666666,"positive_recall":0.94,"specificity":0.8533333333333334,"precision":0.8650306748466258,"balanced_accuracy":0.8966666666666667,"f1":0.9009584664536742,"true_positive":141,"false_positive":22,"false_negative":9,"true_negative":128,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6702441216832,"baseline_adjusted_skill":0.7933333333333334,"baseline_adjusted_score":0.7933333333333334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-knowledge-hard-choice · target_mass_at_prediction	74.67%	Setup & source S1MB datasets/generalization-japanese-knowledge-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.7466666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.6622222222222223,"baseline_adjusted_score":0.6622222222222223}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-knowledge-hard-noul · binary_brier	0.2081 Brier	Setup & source S1MB datasets/generalization-japanese-knowledge-hard-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.20811235454495333,"accuracy":0.6466666666666666,"positive_recall":0.86,"specificity":0.43333333333333335,"precision":0.602803738317757,"balanced_accuracy":0.6466666666666667,"f1":0.7087912087912088,"true_positive":129,"false_positive":85,"false_negative":21,"true_negative":65,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.16755058182018667,"baseline_adjusted_skill":0.29333333333333345,"baseline_adjusted_score":0.29333333333333345}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-knowledge-hard-score · normalized_expected_score_mae	0.3203 MAE	Setup & source S1MB datasets/generalization-japanese-knowledge-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.32025149024248856,"normalized_score_rmse":0.3740426478900636,"constant_score_baseline":0.5,"constant_mae_baseline":0.3333333333333333,"baseline_adjusted_skill":0.03924552927253433,"baseline_adjusted_score":0.03924552927253433}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-math-formal-hard-choice · target_mass_at_prediction	34.67%	Setup & source S1MB datasets/generalization-japanese-math-formal-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.3466666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.12888888888888891,"baseline_adjusted_score":0.12888888888888891}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-math-formal-hard-noul · binary_brier	0.2546 Brier	Setup & source S1MB datasets/generalization-japanese-math-formal-hard-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.25459368391493004,"accuracy":0.5066666666666667,"positive_recall":0.4,"specificity":0.6133333333333333,"precision":0.5084745762711864,"balanced_accuracy":0.5066666666666666,"f1":0.44776119402985076,"true_positive":60,"false_positive":58,"false_negative":90,"true_negative":92,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.018374735659720143,"baseline_adjusted_skill":0.013333333333333197,"baseline_adjusted_score":0.013333333333333197}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-japanese-math-formal-hard-score · normalized_expected_score_mae	0.3097 MAE	Setup & source S1MB datasets/generalization-japanese-math-formal-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.3096516674578692,"normalized_score_rmse":0.3613421042495995,"constant_score_baseline":0.5,"constant_mae_baseline":0.3,"baseline_adjusted_skill":-0.03217222485956417,"baseline_adjusted_score":0}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-easy-choice · target_mass_at_prediction	88.00%	Setup & source S1MB datasets/generalization-vision-easy-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3,"baseline_adjusted_skill":0.8285714285714287,"baseline_adjusted_score":0.8285714285714287}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-easy-noul · binary_brier	0.1099 Brier	Setup & source S1MB datasets/generalization-vision-easy-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10987046062283667,"accuracy":0.8666666666666667,"positive_recall":0.8993288590604027,"specificity":0.8344370860927153,"precision":0.8427672955974843,"balanced_accuracy":0.866882972576559,"f1":0.8701298701298701,"true_positive":134,"false_positive":25,"false_negative":15,"true_negative":126,"positive_prevalence":0.49666666666666665,"majority_accuracy_baseline":0.5033333333333334,"constant_brier_baseline":0.2499888888888889,"brier_skill":0.5604986241141695,"baseline_adjusted_skill":0.733765945153118,"baseline_adjusted_score":0.733765945153118}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-easy-score · normalized_expected_score_mae	0.2995 MAE	Setup & source S1MB datasets/generalization-vision-easy-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.2994853306734164,"normalized_score_rmse":0.3600184005496489,"constant_score_baseline":0.5,"constant_mae_baseline":0.36666666666666664,"baseline_adjusted_skill":0.18322182543613708,"baseline_adjusted_score":0.18322182543613708}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-hard-choice · target_mass_at_prediction	40.00%	Setup & source S1MB datasets/generalization-vision-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.4,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.2866666666666667,"baseline_adjusted_skill":0.1588785046728972,"baseline_adjusted_score":0.1588785046728972}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-hard-noul · binary_brier	0.2793 Brier	Setup & source S1MB datasets/generalization-vision-hard-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.27934352160268666,"accuracy":0.5166666666666667,"positive_recall":0.6878980891719745,"specificity":0.32867132867132864,"precision":0.5294117647058824,"balanced_accuracy":0.5082847089216516,"f1":0.5983379501385041,"true_positive":108,"false_positive":96,"false_negative":49,"true_negative":47,"positive_prevalence":0.5233333333333333,"majority_accuracy_baseline":0.5233333333333333,"constant_brier_baseline":0.24945555555555557,"brier_skill":-0.11981278981968724,"baseline_adjusted_skill":0.016569417843303125,"baseline_adjusted_score":0.016569417843303125}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · generalization-vision-hard-score · normalized_expected_score_mae	0.3147 MAE	Setup & source S1MB datasets/generalization-vision-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.3146685745447347,"normalized_score_rmse":0.3785282171605838,"constant_score_baseline":0.5,"constant_mae_baseline":0.3016666666666667,"baseline_adjusted_skill":-0.04310024710961757,"baseline_adjusted_score":0}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · go-emotions-noul-test-v1 · binary_brier	0.1113 Brier	Setup & source S1MB datasets/go_emotions / test · binary_brier. Population 100 cases / 2800 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.11127623209841675,"accuracy":0.8144464285714285,"positive_recall":0.6322785437952585,"specificity":0.8260753930812631,"precision":0.18835697399527188,"balanced_accuracy":0.7291769684382607,"f1":0.29024840053732837,"true_positive":106.23333333333333,"false_positive":457.76666666666665,"false_negative":61.78333333333333,"true_negative":2174.2166666666667,"positive_prevalence":0.060005952380952375,"majority_accuracy_baseline":0.9399940476190476,"constant_brier_baseline":0.03220097218679138,"brier_skill":-2.455679271201057,"baseline_adjusted_skill":0.4583539368765215,"baseline_adjusted_score":0.4583539368765215}. Counts: {"cases": 100, "expected": 2800, "succeeded": 2800, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · gpqa-ja-main-choice · target_mass_at_prediction	26.67%	Setup & source S1MB datasets/gpqa-ja-main-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.26666666666666666,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.022222222222222216,"baseline_adjusted_score":0.022222222222222216}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · gretel-pii-choice-test-v1 · target_mass_at_prediction	58.64%	Setup & source S1MB datasets/gretel_pii / test · target_mass_at_prediction. Population 100 cases / 353 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.5864022662889519,"uniform_accuracy_baseline":0.027777777777777776,"fixed_answer_accuracy_baseline":0.1274787535410765,"baseline_adjusted_skill":0.525974025974026,"baseline_adjusted_score":0.525974025974026}. Counts: {"cases": 100, "expected": 353, "succeeded": 353, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · gsm8k-choice-test-v1 · target_mass_at_prediction	61.00%	Setup & source S1MB datasets/gsm8k / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.61,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3,"baseline_adjusted_skill":0.4428571428571429,"baseline_adjusted_score":0.4428571428571429}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · hans-noul-test-v1 · binary_brier	0.1623 Brier	Setup & source S1MB datasets/hans / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16234406311113003,"accuracy":0.73,"positive_recall":0.96,"specificity":0.5,"precision":0.6575342465753424,"balanced_accuracy":0.73,"f1":0.7804878048780488,"true_positive":48,"false_positive":25,"false_negative":2,"true_negative":25,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.35062374755547987,"baseline_adjusted_skill":0.45999999999999996,"baseline_adjusted_score":0.45999999999999996}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · hatecheck-noul-test-v1 · binary_brier	0.2002 Brier	Setup & source S1MB datasets/hatecheck / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.20020892741123,"accuracy":0.69,"positive_recall":0.9,"specificity":0.48,"precision":0.6338028169014085,"balanced_accuracy":0.69,"f1":0.743801652892562,"true_positive":45,"false_positive":26,"false_negative":5,"true_negative":24,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.19916429035508,"baseline_adjusted_skill":0.3799999999999999,"baseline_adjusted_score":0.3799999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · hh-rlhf-choice-test-v1 · target_mass_at_prediction	57.00%	Setup & source S1MB datasets/hh_rlhf / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.57,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.61,"baseline_adjusted_skill":-0.10256410256410266,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · hwu64-choice-test-v1 · target_mass_at_prediction	82.00%	Setup & source S1MB datasets/hwu64 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.82,"uniform_accuracy_baseline":0.015625,"fixed_answer_accuracy_baseline":0.06,"baseline_adjusted_skill":0.8085106382978724,"baseline_adjusted_score":0.8085106382978724}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · impli-noul-test-v1 · binary_brier	0.1835 Brier	Setup & source S1MB datasets/impli / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.18352758088047,"accuracy":0.7,"positive_recall":0.86,"specificity":0.54,"precision":0.6515151515151515,"balanced_accuracy":0.7,"f1":0.7413793103448276,"true_positive":43,"false_positive":23,"false_negative":7,"true_negative":27,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.26588967647812,"baseline_adjusted_skill":0.3999999999999999,"baseline_adjusted_score":0.3999999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jamc-qa-choice · target_mass_at_prediction	29.67%	Setup & source S1MB datasets/jamc-qa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.2966666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.062222222222222255,"baseline_adjusted_score":0.062222222222222255}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jamp-choice · target_mass_at_prediction	54.33%	Setup & source S1MB datasets/jamp-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.5433333333333333,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.315,"baseline_adjusted_score":0.315}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · janli-choice · target_mass_at_prediction	57.19%	Setup & source S1MB datasets/janli-choice / test · target_mass_at_prediction. Population 285 cases / 285 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.5719298245614035,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5263157894736842,"baseline_adjusted_skill":0.09629629629629635,"baseline_adjusted_score":0.09629629629629635}. Counts: {"cases": 285, "expected": 285, "succeeded": 285, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 285 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jcola-noul · binary_brier	0.2552 Brier	Setup & source S1MB datasets/jcola-noul / test · binary_brier. Population 299 cases / 299 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.2552081150100836,"accuracy":0.5083612040133779,"positive_recall":0.8133333333333334,"specificity":0.20134228187919462,"precision":0.5062240663900415,"balanced_accuracy":0.507337807606264,"f1":0.6240409207161125,"true_positive":122,"false_positive":119,"false_negative":28,"true_negative":30,"positive_prevalence":0.5016722408026756,"majority_accuracy_baseline":0.5016722408026756,"constant_brier_baseline":0.24999720361069788,"brier_skill":-0.020843878747941025,"baseline_adjusted_skill":0.014675615212528026,"baseline_adjusted_score":0.014675615212528026}. Counts: {"cases": 299, "expected": 299, "succeeded": 299, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 299 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jcommonsenseqa-choice · target_mass_at_prediction	75.67%	Setup & source S1MB datasets/jcommonsenseqa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.7566666666666667,"uniform_accuracy_baseline":0.2,"fixed_answer_accuracy_baseline":0.2,"baseline_adjusted_skill":0.6958333333333333,"baseline_adjusted_score":0.6958333333333333}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jculture-mcq-noul · binary_brier	0.3519 Brier	Setup & source S1MB datasets/jculture-mcq-noul / test · binary_brier. Population 283 cases / 283 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.3519059107110389,"accuracy":0.4911660777385159,"positive_recall":0.7819548872180451,"specificity":0.23333333333333334,"precision":0.4748858447488584,"balanced_accuracy":0.5076441102756892,"f1":0.5909090909090909,"true_positive":104,"false_positive":115,"false_negative":29,"true_negative":35,"positive_prevalence":0.46996466431095407,"majority_accuracy_baseline":0.5300353356890459,"constant_brier_baseline":0.24909787861004634,"brier_skill":-0.41272142771610976,"baseline_adjusted_skill":0.015288220551378462,"baseline_adjusted_score":0.015288220551378462}. Counts: {"cases": 283, "expected": 283, "succeeded": 283, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 283 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jmmlu-japan-choice · target_mass_at_prediction	52.67%	Setup & source S1MB datasets/jmmlu-japan-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.5266666666666666,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.3688888888888888,"baseline_adjusted_score":0.3688888888888888}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jmteb-retrieval · target_mass_at_prediction	92.00%	Setup & source S1MB datasets/jmteb-retrieval / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.92,"uniform_accuracy_baseline":0.20616666666666666,"fixed_answer_accuracy_baseline":0.21333333333333335,"baseline_adjusted_skill":0.8983050847457628,"baseline_adjusted_score":0.8983050847457628}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jnli-choice · target_mass_at_prediction	69.00%	Setup & source S1MB datasets/jnli-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.69,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.5349999999999999,"baseline_adjusted_score":0.5349999999999999}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jsem-choice · target_mass_at_prediction	51.00%	Setup & source S1MB datasets/jsem-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.51,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.37333333333333335,"baseline_adjusted_skill":0.2180851063829787,"baseline_adjusted_score":0.2180851063829787}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jsick-sts-score · normalized_expected_score_mae	0.1649 MAE	Setup & source S1MB datasets/jsick-sts-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.16488168227454508,"normalized_score_rmse":0.1919794217074036,"constant_score_baseline":0.5,"constant_mae_baseline":0.26208333333333333,"baseline_adjusted_skill":0.370880703562944,"baseline_adjusted_score":0.370880703562944}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · jsts-score · normalized_expected_score_mae	0.1374 MAE	Setup & source S1MB datasets/jsts-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.13735466698075918,"normalized_score_rmse":0.1707946283104116,"constant_score_baseline":0.52,"constant_mae_baseline":0.24733333333333335,"baseline_adjusted_skill":0.44465768067078504,"baseline_adjusted_score":0.44465768067078504}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · koniq-score · normalized_expected_score_mae	0.1267 MAE	Setup & source S1MB datasets/koniq-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.12669309789718325,"normalized_score_rmse":0.14419214300880967,"constant_score_baseline":0.578319964349225,"constant_mae_baseline":0.11212962028602311,"baseline_adjusted_skill":-0.12988073600901573,"baseline_adjusted_score":0}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · kuci-choice · target_mass_at_prediction	56.00%	Setup & source S1MB datasets/kuci-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.56,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.4133333333333334,"baseline_adjusted_score":0.4133333333333334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · laya--ag-news-choice-test-v1 · target_mass_at_prediction	88.00%	Setup & source S1MB datasets/laya__ag_news / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.33,"baseline_adjusted_skill":0.8208955223880599,"baseline_adjusted_score":0.8208955223880599}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · laya--enron-spam-noul-test-v1 · binary_brier	0.1471 Brier	Setup & source S1MB datasets/laya__enron_spam / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.14710850909392,"accuracy":0.79,"positive_recall":0.66,"specificity":0.92,"precision":0.8918918918918919,"balanced_accuracy":0.79,"f1":0.7586206896551724,"true_positive":33,"false_positive":4,"false_negative":17,"true_negative":46,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.41156596362431996,"baseline_adjusted_skill":0.5800000000000001,"baseline_adjusted_score":0.5800000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · laya--typed-decisions-choice-test-v1 · target_mass_at_prediction	41.65%	Setup & source S1MB datasets/laya__typed_decisions / test · target_mass_at_prediction. Population 100 cases / 148 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.4165445202702703,"uniform_accuracy_baseline":0.23445945945945948,"fixed_answer_accuracy_baseline":0.2623439054054054,"baseline_adjusted_skill":0.2090413351083493,"baseline_adjusted_score":0.2090413351083493}. Counts: {"cases": 100, "expected": 148, "succeeded": 148, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · laya--typed-decisions-noul-test-v1 · binary_brier	0.0681 Brier	Setup & source S1MB datasets/laya__typed_decisions / test · binary_brier. Population 100 cases / 152 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.06807214165176316,"accuracy":0.6470175460526315,"positive_recall":0.7383778363213716,"specificity":0.5600222666332751,"precision":0.6150936516853933,"balanced_accuracy":0.6492000514773233,"f1":0.6711209267294178,"true_positive":54.743335,"false_positive":34.256665,"false_negative":19.396668,"true_negative":43.603332,"positive_prevalence":0.4877631776315789,"majority_accuracy_baseline":0.512236822368421,"constant_brier_baseline":0.09451910039768556,"brier_skill":0.279805442864435,"baseline_adjusted_skill":0.2984001029546466,"baseline_adjusted_score":0.2984001029546466}. Counts: {"cases": 100, "expected": 152, "succeeded": 152, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · laya--typed-decisions-score-test-v1 · normalized_expected_score_mae	0.1596 MAE	Setup & source S1MB datasets/laya__typed_decisions / test · normalized_expected_score_mae. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.1596497086858443,"normalized_score_rmse":0.2300668614024061,"constant_score_baseline":0.5841668333333333,"constant_mae_baseline":0.21700552708333334,"baseline_adjusted_skill":0.26430579519508535,"baseline_adjusted_score":0.26430579519508535}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · lexcomp-noul-test-v1 · binary_brier	0.2976 Brier	Setup & source S1MB datasets/lexcomp / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.29758191075789997,"accuracy":0.54,"positive_recall":0.88,"specificity":0.2,"precision":0.5238095238095238,"balanced_accuracy":0.54,"f1":0.6567164179104478,"true_positive":44,"false_positive":40,"false_negative":6,"true_negative":10,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.19032764303159988,"baseline_adjusted_skill":0.08000000000000007,"baseline_adjusted_score":0.08000000000000007}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · logical-entailment-noul-test-v1 · binary_brier	0.2957 Brier	Setup & source S1MB datasets/logical_entailment / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.29568702495358,"accuracy":0.52,"positive_recall":0.04,"specificity":1,"precision":1,"balanced_accuracy":0.52,"f1":0.07692307692307693,"true_positive":2,"false_positive":0,"false_negative":48,"true_negative":50,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.18274809981432005,"baseline_adjusted_skill":0.040000000000000036,"baseline_adjusted_score":0.040000000000000036}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · lonli-choice-test-v1 · target_mass_at_prediction	61.00%	Setup & source S1MB datasets/lonli / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.61,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.39,"baseline_adjusted_skill":0.360655737704918,"baseline_adjusted_score":0.360655737704918}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · massive-choice-test-v1 · target_mass_at_prediction	86.00%	Setup & source S1MB datasets/massive / test · target_mass_at_prediction. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.86,"uniform_accuracy_baseline":0.036111111111111115,"fixed_answer_accuracy_baseline":0.08833333333333333,"baseline_adjusted_skill":0.8464351005484461,"baseline_adjusted_score":0.8464351005484461}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · mintaka-ja-boolean-noul · binary_brier	0.2719 Brier	Setup & source S1MB datasets/mintaka-ja-boolean-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.27185915489335666,"accuracy":0.53,"positive_recall":0.6,"specificity":0.46,"precision":0.5263157894736842,"balanced_accuracy":0.53,"f1":0.5607476635514018,"true_positive":90,"false_positive":81,"false_negative":60,"true_negative":69,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.08743661957342663,"baseline_adjusted_skill":0.06000000000000005,"baseline_adjusted_score":0.06000000000000005}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · miqa-choice-test-v1 · target_mass_at_prediction	88.00%	Setup & source S1MB datasets/miqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.56,"baseline_adjusted_skill":0.7272727272727273,"baseline_adjusted_score":0.7272727272727273}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · mmlu-prox-ja-choice · target_mass_at_prediction	22.67%	Setup & source S1MB datasets/mmlu-prox-ja-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.22666666666666666,"uniform_accuracy_baseline":0.11021428571428572,"fixed_answer_accuracy_baseline":0.12554761904761905,"baseline_adjusted_skill":0.11563699730443541,"baseline_adjusted_score":0.11563699730443541}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · mmmlu-choice · target_mass_at_prediction	38.00%	Setup & source S1MB datasets/mmmlu-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.38,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.17333333333333334,"baseline_adjusted_score":0.17333333333333334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · mtop-choice-test-v1 · target_mass_at_prediction	81.50%	Setup & source S1MB datasets/mtop / test · target_mass_at_prediction. Population 100 cases / 200 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.815,"uniform_accuracy_baseline":0.049879324215607396,"fixed_answer_accuracy_baseline":0.12442477876106194,"baseline_adjusted_skill":0.7887103294926218,"baseline_adjusted_score":0.7887103294926218}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · mvtec-ad-noul · binary_brier	0.2671 Brier	Setup & source S1MB datasets/mvtec-ad-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.26712196801539334,"accuracy":0.5466666666666666,"positive_recall":0.4398148148148148,"specificity":0.8214285714285714,"precision":0.8636363636363636,"balanced_accuracy":0.6306216931216931,"f1":0.5828220858895705,"true_positive":95,"false_positive":15,"false_negative":121,"true_negative":69,"positive_prevalence":0.72,"majority_accuracy_baseline":0.72,"constant_brier_baseline":0.2016,"brier_skill":-0.32500976198111786,"baseline_adjusted_skill":0.2612433862433863,"baseline_adjusted_score":0.2612433862433863}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · nights-choice · target_mass_at_prediction	86.67%	Setup & source S1MB datasets/nights-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8666666666666667,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5033333333333333,"baseline_adjusted_skill":0.7315436241610739,"baseline_adjusted_score":0.7315436241610739}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · nlsat-noul-test-v1 · binary_brier	0.2709 Brier	Setup & source S1MB datasets/nlsat / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.27089385346715,"accuracy":0.5,"positive_recall":0,"specificity":1,"precision":null,"balanced_accuracy":0.5,"f1":0,"true_positive":0,"false_positive":0,"false_negative":50,"true_negative":50,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.08357541386860001,"baseline_adjusted_skill":0,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · nlvr2-noul · binary_brier	0.2351 Brier	Setup & source S1MB datasets/nlvr2-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.23513984160534998,"accuracy":0.6133333333333333,"positive_recall":0.45625,"specificity":0.7928571428571428,"precision":0.7156862745098039,"balanced_accuracy":0.6245535714285714,"f1":0.5572519083969466,"true_positive":73,"false_positive":29,"false_negative":87,"true_negative":111,"positive_prevalence":0.5333333333333333,"majority_accuracy_baseline":0.5333333333333333,"constant_brier_baseline":0.2488888888888889,"brier_skill":0.055241707835647524,"baseline_adjusted_skill":0.24910714285714275,"baseline_adjusted_score":0.24910714285714275}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--amount-extraction-control-v1-choice-test-v1 · target_mass_at_prediction	80.56%	Setup & source S1MB datasets/open_jev__amount-extraction-control-v1 / test · target_mass_at_prediction. Population 36 cases / 36 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8055555555555556,"uniform_accuracy_baseline":0.23888888888888887,"fixed_answer_accuracy_baseline":0.2916666666666667,"baseline_adjusted_skill":0.7254901960784313,"baseline_adjusted_score":0.7254901960784313}. Counts: {"cases": 36, "expected": 36, "succeeded": 36, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 36 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--amount-extraction-control-v1-noul-test-v1 · binary_brier	0.2464 Brier	Setup & source S1MB datasets/open_jev__amount-extraction-control-v1 / test · binary_brier. Population 64 cases / 64 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.24644626098896874,"accuracy":0.53125,"positive_recall":0.59375,"specificity":0.46875,"precision":0.5277777777777778,"balanced_accuracy":0.53125,"f1":0.5588235294117647,"true_positive":19,"false_positive":17,"false_negative":13,"true_negative":15,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.014214956044125038,"baseline_adjusted_skill":0.0625,"baseline_adjusted_score":0.0625}. Counts: {"cases": 64, "expected": 64, "succeeded": 64, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 64 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--browser-control-v1-choice-test-v1 · target_mass_at_prediction	86.00%	Setup & source S1MB datasets/open_jev__browser-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.86,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":0.6956521739130435,"baseline_adjusted_score":0.6956521739130435}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--citation-control-v1-choice-test-v1 · target_mass_at_prediction	64.00%	Setup & source S1MB datasets/open_jev__citation-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.64,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.34,"baseline_adjusted_skill":0.4545454545454546,"baseline_adjusted_score":0.4545454545454546}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--context-retention-control-v1-noul-test-v1 · binary_brier	0.2307 Brier	Setup & source S1MB datasets/open_jev__context-retention-control-v1 / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.23071961715874997,"accuracy":0.63,"positive_recall":0.54,"specificity":0.72,"precision":0.6585365853658537,"balanced_accuracy":0.63,"f1":0.5934065934065934,"true_positive":27,"false_positive":14,"false_negative":23,"true_negative":36,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.0771215313650001,"baseline_adjusted_skill":0.26,"baseline_adjusted_score":0.26}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--customer-control-v1-choice-test-v1 · target_mass_at_prediction	82.35%	Setup & source S1MB datasets/open_jev__customer-control-v1 / test · target_mass_at_prediction. Population 17 cases / 17 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8235294117647058,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3235294117647059,"baseline_adjusted_skill":0.7391304347826085,"baseline_adjusted_score":0.7391304347826085}. Counts: {"cases": 17, "expected": 17, "succeeded": 17, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 17 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--customer-control-v1-noul-test-v1 · binary_brier	0.1686 Brier	Setup & source S1MB datasets/open_jev__customer-control-v1 / test · binary_brier. Population 33 cases / 33 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16864783477327275,"accuracy":0.7575757575757576,"positive_recall":0.5,"specificity":1,"precision":1,"balanced_accuracy":0.75,"f1":0.6666666666666666,"true_positive":8,"false_positive":0,"false_negative":8,"true_negative":17,"positive_prevalence":0.48484848484848486,"majority_accuracy_baseline":0.5151515151515151,"constant_brier_baseline":0.2497704315886134,"brier_skill":0.32478863210259545,"baseline_adjusted_skill":0.5,"baseline_adjusted_score":0.5}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 33 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--customer-control-v1-score-test-v1 · normalized_expected_score_mae	0.1655 MAE	Setup & source S1MB datasets/open_jev__customer-control-v1 / test · normalized_expected_score_mae. Population 50 cases / 50 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.16545451699539654,"normalized_score_rmse":0.24578402421833692,"constant_score_baseline":0.25,"constant_mae_baseline":0.33,"baseline_adjusted_skill":0.49862267577152564,"baseline_adjusted_score":0.49862267577152564}. Counts: {"cases": 50, "expected": 50, "succeeded": 50, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 50 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--drone-control-v1-choice-test-v1 · target_mass_at_prediction	0.00%	Setup & source S1MB datasets/open_jev__drone-control-v1 / test · target_mass_at_prediction. Population 32 cases / 32 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0,"uniform_accuracy_baseline":0.178125,"fixed_answer_accuracy_baseline":0.2875,"baseline_adjusted_skill":-0.40350877192982454,"baseline_adjusted_score":0}. Counts: {"cases": 32, "expected": 32, "succeeded": 32, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 32 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--drone-control-v1-noul-test-v1 · binary_brier	0.1061 Brier	Setup & source S1MB datasets/open_jev__drone-control-v1 / test · binary_brier. Population 34 cases / 34 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10609851660214706,"accuracy":0.9705882352941176,"positive_recall":1,"specificity":0.9411764705882353,"precision":0.9444444444444444,"balanced_accuracy":0.9705882352941176,"f1":0.9714285714285714,"true_positive":17,"false_positive":1,"false_negative":0,"true_negative":16,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.5756059335914118,"baseline_adjusted_skill":0.9411764705882353,"baseline_adjusted_score":0.9411764705882353}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 34 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--drone-control-v1-score-test-v1 · normalized_expected_score_mae	0.1891 MAE	Setup & source S1MB datasets/open_jev__drone-control-v1 / test · normalized_expected_score_mae. Population 34 cases / 34 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.18909829375217066,"normalized_score_rmse":0.2835178561048283,"constant_score_baseline":0.5,"constant_mae_baseline":0.14705882352941177,"baseline_adjusted_skill":-0.28586839751476045,"baseline_adjusted_score":0}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 34 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--email-selection-control-v1-choice-test-v1 · target_mass_at_prediction	10.42%	Setup & source S1MB datasets/open_jev__email-selection-control-v1 / test · target_mass_at_prediction. Population 48 cases / 48 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.10416666666666667,"uniform_accuracy_baseline":0.13194444444444445,"fixed_answer_accuracy_baseline":0.6006944444444444,"baseline_adjusted_skill":-1.243478260869565,"baseline_adjusted_score":0}. Counts: {"cases": 48, "expected": 48, "succeeded": 48, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 48 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--email-selection-control-v1-noul-test-v1 · binary_brier	0.2487 Brier	Setup & source S1MB datasets/open_jev__email-selection-control-v1 / test · binary_brier. Population 52 cases / 52 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.24868820665657693,"accuracy":0.6923076923076923,"positive_recall":1,"specificity":0.38461538461538464,"precision":0.6190476190476191,"balanced_accuracy":0.6923076923076923,"f1":0.7647058823529411,"true_positive":26,"false_positive":16,"false_negative":0,"true_negative":10,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.005247173373692293,"baseline_adjusted_skill":0.3846153846153846,"baseline_adjusted_score":0.3846153846153846}. Counts: {"cases": 52, "expected": 52, "succeeded": 52, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 52 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--entity-alignment-control-v1-noul-test-v1 · binary_brier	0.2737 Brier	Setup & source S1MB datasets/open_jev__entity-alignment-control-v1 / test · binary_brier. Population 75 cases / 75 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.27373212249278667,"accuracy":0.5066666666666667,"positive_recall":0.23076923076923078,"specificity":0.8055555555555556,"precision":0.5625,"balanced_accuracy":0.5181623931623932,"f1":0.32727272727272727,"true_positive":9,"false_positive":7,"false_negative":30,"true_negative":29,"positive_prevalence":0.52,"majority_accuracy_baseline":0.52,"constant_brier_baseline":0.24960000000000004,"brier_skill":-0.09668318306404888,"baseline_adjusted_skill":0.03632478632478642,"baseline_adjusted_score":0.03632478632478642}. Counts: {"cases": 75, "expected": 75, "succeeded": 75, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 75 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--entity-alignment-control-v1-score-test-v1 · normalized_expected_score_mae	0.3319 MAE	Setup & source S1MB datasets/open_jev__entity-alignment-control-v1 / test · normalized_expected_score_mae. Population 25 cases / 25 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.33188339636967945,"normalized_score_rmse":0.4031282362280898,"constant_score_baseline":0.5,"constant_mae_baseline":0.28,"baseline_adjusted_skill":-0.18529784417742645,"baseline_adjusted_score":0}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 25 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--ir-control-v1-choice-test-v1 · target_mass_at_prediction	47.06%	Setup & source S1MB datasets/open_jev__ir-control-v1 / test · target_mass_at_prediction. Population 17 cases / 17 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.47058823529411764,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.2647058823529412,"baseline_adjusted_skill":0.27999999999999997,"baseline_adjusted_score":0.27999999999999997}. Counts: {"cases": 17, "expected": 17, "succeeded": 17, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 17 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--ir-control-v1-noul-test-v1 · binary_brier	0.1502 Brier	Setup & source S1MB datasets/open_jev__ir-control-v1 / test · binary_brier. Population 28 cases / 28 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.15022999315517857,"accuracy":0.8214285714285714,"positive_recall":1,"specificity":0.6428571428571429,"precision":0.7368421052631579,"balanced_accuracy":0.8214285714285714,"f1":0.8484848484848485,"true_positive":14,"false_positive":5,"false_negative":0,"true_negative":9,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.3990800273792857,"baseline_adjusted_skill":0.6428571428571428,"baseline_adjusted_score":0.6428571428571428}. Counts: {"cases": 28, "expected": 28, "succeeded": 28, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 28 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--ir-control-v1-score-test-v1 · normalized_expected_score_mae	0.2395 MAE	Setup & source S1MB datasets/open_jev__ir-control-v1 / test · normalized_expected_score_mae. Population 55 cases / 55 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.23950980952980114,"normalized_score_rmse":0.30817198135718665,"constant_score_baseline":0.3333333333333333,"constant_mae_baseline":0.1696969696969697,"baseline_adjusted_skill":-0.41139709187204243,"baseline_adjusted_score":0}. Counts: {"cases": 55, "expected": 55, "succeeded": 55, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 55 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--mailroom-control-v1-choice-test-v1 · target_mass_at_prediction	93.33%	Setup & source S1MB datasets/open_jev__mailroom-control-v1 / test · target_mass_at_prediction. Population 15 cases / 15 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.9333333333333333,"uniform_accuracy_baseline":0.15555555555555553,"fixed_answer_accuracy_baseline":0.2,"baseline_adjusted_skill":0.9166666666666667,"baseline_adjusted_score":0.9166666666666667}. Counts: {"cases": 15, "expected": 15, "succeeded": 15, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 15 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--mailroom-control-v1-noul-test-v1 · binary_brier	0.1001 Brier	Setup & source S1MB datasets/open_jev__mailroom-control-v1 / test · binary_brier. Population 85 cases / 85 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10005264318182353,"accuracy":0.8588235294117647,"positive_recall":0.9111111111111111,"specificity":0.8,"precision":0.8367346938775511,"balanced_accuracy":0.8555555555555556,"f1":0.8723404255319149,"true_positive":41,"false_positive":8,"false_negative":4,"true_negative":32,"positive_prevalence":0.5294117647058824,"majority_accuracy_baseline":0.5294117647058824,"constant_brier_baseline":0.2491349480968858,"brier_skill":0.5983998072285139,"baseline_adjusted_skill":0.7111111111111112,"baseline_adjusted_score":0.7111111111111112}. Counts: {"cases": 85, "expected": 85, "succeeded": 85, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 85 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--painting-geometry-v1-choice-test-v1 · target_mass_at_prediction	44.00%	Setup & source S1MB datasets/open_jev__painting-geometry-v1 / test · target_mass_at_prediction. Population 25 cases / 25 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.44,"uniform_accuracy_baseline":0.07805555555555556,"fixed_answer_accuracy_baseline":0.24,"baseline_adjusted_skill":0.2631578947368421,"baseline_adjusted_score":0.2631578947368421}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 25 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--painting-geometry-v1-noul-test-v1 · binary_brier	0.2122 Brier	Setup & source S1MB datasets/open_jev__painting-geometry-v1 / test · binary_brier. Population 50 cases / 50 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.21215101772248,"accuracy":0.62,"positive_recall":0.6285714285714286,"specificity":0.6,"precision":0.7857142857142857,"balanced_accuracy":0.6142857142857143,"f1":0.6984126984126984,"true_positive":22,"false_positive":6,"false_negative":13,"true_negative":9,"positive_prevalence":0.7,"majority_accuracy_baseline":0.7,"constant_brier_baseline":0.21,"brier_skill":-0.01024294153561911,"baseline_adjusted_skill":0.22857142857142865,"baseline_adjusted_score":0.22857142857142865}. Counts: {"cases": 50, "expected": 50, "succeeded": 50, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 50 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--painting-geometry-v1-score-test-v1 · normalized_expected_score_mae	0.3382 MAE	Setup & source S1MB datasets/open_jev__painting-geometry-v1 / test · normalized_expected_score_mae. Population 25 cases / 25 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.33821660206704024,"normalized_score_rmse":0.43044384909102784,"constant_score_baseline":0.5,"constant_mae_baseline":0.28,"baseline_adjusted_skill":-0.20791643595371512,"baseline_adjusted_score":0}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 25 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--phone-extraction-control-v1-choice-test-v1 · target_mass_at_prediction	62.03%	Setup & source S1MB datasets/open_jev__phone-extraction-control-v1 / test · target_mass_at_prediction. Population 79 cases / 79 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.620253164556962,"uniform_accuracy_baseline":0.24240506329113923,"fixed_answer_accuracy_baseline":0.4177215189873418,"baseline_adjusted_skill":0.3478260869565217,"baseline_adjusted_score":0.3478260869565217}. Counts: {"cases": 79, "expected": 79, "succeeded": 79, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 79 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--phone-extraction-control-v1-noul-test-v1 · binary_brier	0.1613 Brier	Setup & source S1MB datasets/open_jev__phone-extraction-control-v1 / test · binary_brier. Population 21 cases / 21 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16125406227947617,"accuracy":0.7619047619047619,"positive_recall":1,"specificity":0.5,"precision":0.6875,"balanced_accuracy":0.75,"f1":0.8148148148148148,"true_positive":11,"false_positive":5,"false_negative":0,"true_negative":5,"positive_prevalence":0.5238095238095238,"majority_accuracy_baseline":0.5238095238095238,"constant_brier_baseline":0.2494331065759637,"brier_skill":0.35351780486137274,"baseline_adjusted_skill":0.5,"baseline_adjusted_score":0.5}. Counts: {"cases": 21, "expected": 21, "succeeded": 21, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 21 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--reasoning-control-v1-choice-test-v1 · target_mass_at_prediction	48.48%	Setup & source S1MB datasets/open_jev__reasoning-control-v1 / test · target_mass_at_prediction. Population 33 cases / 33 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.48484848484848486,"uniform_accuracy_baseline":0.32373737373737377,"fixed_answer_accuracy_baseline":0.3939393939393939,"baseline_adjusted_skill":0.15000000000000005,"baseline_adjusted_score":0.15000000000000005}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 33 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--reasoning-control-v1-noul-test-v1 · binary_brier	0.2102 Brier	Setup & source S1MB datasets/open_jev__reasoning-control-v1 / test · binary_brier. Population 33 cases / 33 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.2101889609770909,"accuracy":0.6666666666666666,"positive_recall":0.7058823529411765,"specificity":0.625,"precision":0.6666666666666666,"balanced_accuracy":0.6654411764705883,"f1":0.6857142857142857,"true_positive":12,"false_positive":6,"false_negative":5,"true_negative":10,"positive_prevalence":0.5151515151515151,"majority_accuracy_baseline":0.5151515151515151,"constant_brier_baseline":0.2497704315886134,"brier_skill":0.15847140255863235,"baseline_adjusted_skill":0.3308823529411766,"baseline_adjusted_score":0.3308823529411766}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 33 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--reasoning-control-v1-score-test-v1 · normalized_expected_score_mae	0.3194 MAE	Setup & source S1MB datasets/open_jev__reasoning-control-v1 / test · normalized_expected_score_mae. Population 34 cases / 34 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.3193532209483868,"normalized_score_rmse":0.3759947353181299,"constant_score_baseline":0.7083333333333333,"constant_mae_baseline":0.3431372549019608,"baseline_adjusted_skill":0.06931347037898705,"baseline_adjusted_score":0.06931347037898705}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 34 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--silent-failure-control-v1-noul-test-v1 · binary_brier	0.3745 Brier	Setup & source S1MB datasets/open_jev__silent-failure-control-v1 / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.37447354064210003,"accuracy":0.32,"positive_recall":0.48,"specificity":0.16,"precision":0.36363636363636365,"balanced_accuracy":0.32,"f1":0.41379310344827586,"true_positive":24,"false_positive":42,"false_negative":26,"true_negative":8,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.49789416256840013,"baseline_adjusted_skill":-0.36,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--snake-v1-choice-test-v1 · target_mass_at_prediction	37.33%	Setup & source S1MB datasets/open_jev__snake-v1 / test · target_mass_at_prediction. Population 25 cases / 25 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.37333333333333335,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.35333333333333333,"baseline_adjusted_skill":0.030927835051546417,"baseline_adjusted_score":0.030927835051546417}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 25 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--snake-v1-noul-test-v1 · binary_brier	0.2540 Brier	Setup & source S1MB datasets/open_jev__snake-v1 / test · binary_brier. Population 75 cases / 75 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.25402951646570665,"accuracy":0.52,"positive_recall":0.02702702702702703,"specificity":1,"precision":1,"balanced_accuracy":0.5135135135135135,"f1":0.05263157894736842,"true_positive":1,"false_positive":0,"false_negative":36,"true_negative":38,"positive_prevalence":0.49333333333333335,"majority_accuracy_baseline":0.5066666666666666,"constant_brier_baseline":0.24995555555555554,"brier_skill":-0.016298741194594513,"baseline_adjusted_skill":0.027027027027026973,"baseline_adjusted_score":0.027027027027026973}. Counts: {"cases": 75, "expected": 75, "succeeded": 75, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 75 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--sponsor-segment-control-v1-choice-test-v1 · target_mass_at_prediction	55.00%	Setup & source S1MB datasets/open_jev__sponsor-segment-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.55,"uniform_accuracy_baseline":0.14285714285714285,"fixed_answer_accuracy_baseline":0.22,"baseline_adjusted_skill":0.42307692307692313,"baseline_adjusted_score":0.42307692307692313}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--tic-tac-toe-v1-choice-test-v1 · target_mass_at_prediction	34.00%	Setup & source S1MB datasets/open_jev__tic_tac_toe-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.34,"uniform_accuracy_baseline":0.32416666666666666,"fixed_answer_accuracy_baseline":0.34,"baseline_adjusted_skill":0,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--vizdoom-basic-v1-noul-test-v1 · binary_brier	0.1604 Brier	Setup & source S1MB datasets/open_jev__vizdoom-basic-v1 / test · binary_brier. Population 33 cases / 33 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16040875696639395,"accuracy":0.9696969696969697,"positive_recall":1,"specificity":0.9411764705882353,"precision":0.9411764705882353,"balanced_accuracy":0.9705882352941176,"f1":0.9696969696969697,"true_positive":16,"false_positive":1,"false_negative":0,"true_negative":16,"positive_prevalence":0.48484848484848486,"majority_accuracy_baseline":0.5151515151515151,"constant_brier_baseline":0.2497704315886134,"brier_skill":0.3577752340573418,"baseline_adjusted_skill":0.9411764705882353,"baseline_adjusted_score":0.9411764705882353}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 33 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--vizdoom-basic-v1-score-test-v1 · normalized_expected_score_mae	0.2216 MAE	Setup & source S1MB datasets/open_jev__vizdoom-basic-v1 / test · normalized_expected_score_mae. Population 33 cases / 33 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.22157090119326223,"normalized_score_rmse":0.2577561240116941,"constant_score_baseline":0.5,"constant_mae_baseline":0.2196969696969697,"baseline_adjusted_skill":-0.008529619224503904,"baseline_adjusted_score":0}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 33 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--workflow-controls-v1--agent-trace-observability-noul-test-v1 · binary_brier	0.1679 Brier	Setup & source S1MB datasets/open_jev__workflow-controls-v1__agent_trace_observability / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16792424674140002,"accuracy":0.77,"positive_recall":0.5454545454545454,"specificity":0.8333333333333334,"precision":0.48,"balanced_accuracy":0.6893939393939394,"f1":0.5106382978723404,"true_positive":12,"false_positive":13,"false_negative":10,"true_negative":65,"positive_prevalence":0.22,"majority_accuracy_baseline":0.78,"constant_brier_baseline":0.1716,"brier_skill":0.0214204735349649,"baseline_adjusted_skill":0.3787878787878789,"baseline_adjusted_score":0.3787878787878789}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--workflow-controls-v1--customer-service-noul-test-v1 · binary_brier	0.1582 Brier	Setup & source S1MB datasets/open_jev__workflow-controls-v1__customer_service / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.1581846166697,"accuracy":0.8,"positive_recall":0.7297297297297297,"specificity":0.8412698412698413,"precision":0.7297297297297297,"balanced_accuracy":0.7854997854997855,"f1":0.7297297297297297,"true_positive":27,"false_positive":10,"false_negative":10,"true_negative":53,"positive_prevalence":0.37,"majority_accuracy_baseline":0.63,"constant_brier_baseline":0.23310000000000003,"brier_skill":0.321387315874303,"baseline_adjusted_skill":0.570999570999571,"baseline_adjusted_score":0.570999570999571}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--workflow-controls-v1--invoice-processing-noul-test-v1 · binary_brier	0.2084 Brier	Setup & source S1MB datasets/open_jev__workflow-controls-v1__invoice_processing / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.20838807423453,"accuracy":0.67,"positive_recall":0.46153846153846156,"specificity":0.8032786885245902,"precision":0.6,"balanced_accuracy":0.6324085750315258,"f1":0.5217391304347826,"true_positive":18,"false_positive":12,"false_negative":21,"true_negative":49,"positive_prevalence":0.39,"majority_accuracy_baseline":0.61,"constant_brier_baseline":0.2379,"brier_skill":0.12405181069974769,"baseline_adjusted_skill":0.2648171500630516,"baseline_adjusted_score":0.2648171500630516}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · open-jev--workflow-controls-v1--security-incidents-noul-test-v1 · binary_brier	0.1742 Brier	Setup & source S1MB datasets/open_jev__workflow-controls-v1__security_incidents / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.17416469508077997,"accuracy":0.68,"positive_recall":0.8709677419354839,"specificity":0.5942028985507246,"precision":0.4909090909090909,"balanced_accuracy":0.7325853202431043,"f1":0.627906976744186,"true_positive":27,"false_positive":28,"false_negative":4,"true_negative":41,"positive_prevalence":0.31,"majority_accuracy_baseline":0.69,"constant_brier_baseline":0.21389999999999998,"brier_skill":0.185765801398878,"baseline_adjusted_skill":0.4651706404862086,"baseline_adjusted_score":0.4651706404862086}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · openbookqa-choice-test-v1 · target_mass_at_prediction	68.00%	Setup & source S1MB datasets/openbookqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.68,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.28,"baseline_adjusted_skill":0.5555555555555556,"baseline_adjusted_score":0.5555555555555556}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · patent-similarity-score-test-v1 · normalized_expected_score_mae	0.2260 MAE	Setup & source S1MB datasets/patent_similarity / test · normalized_expected_score_mae. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.22603835586767143,"normalized_score_rmse":0.27569082625586233,"constant_score_baseline":0.25,"constant_mae_baseline":0.2325,"baseline_adjusted_skill":0.0277920177734563,"baseline_adjusted_score":0.0277920177734563}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · paws-noul-test-v1 · binary_brier	0.1599 Brier	Setup & source S1MB datasets/paws / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.15987922038184,"accuracy":0.77,"positive_recall":0.9,"specificity":0.64,"precision":0.7142857142857143,"balanced_accuracy":0.77,"f1":0.7964601769911505,"true_positive":45,"false_positive":18,"false_negative":5,"true_negative":32,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.36048311847264003,"baseline_adjusted_skill":0.54,"baseline_adjusted_score":0.54}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · plane-noul-test-v1 · binary_brier	0.2300 Brier	Setup & source S1MB datasets/plane / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.23000626744737002,"accuracy":0.58,"positive_recall":0.5,"specificity":0.66,"precision":0.5952380952380952,"balanced_accuracy":0.5800000000000001,"f1":0.5434782608695652,"true_positive":25,"false_positive":17,"false_negative":25,"true_negative":33,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.07997493021051993,"baseline_adjusted_skill":0.16000000000000014,"baseline_adjusted_score":0.16000000000000014}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · poem-sentiment-choice-test-v1 · target_mass_at_prediction	51.00%	Setup & source S1MB datasets/poem_sentiment / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.51,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.65,"baseline_adjusted_skill":-0.4000000000000001,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · pope-noul · binary_brier	0.1225 Brier	Setup & source S1MB datasets/pope-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.12245648992879665,"accuracy":0.8366666666666667,"positive_recall":0.74,"specificity":0.9333333333333333,"precision":0.9173553719008265,"balanced_accuracy":0.8366666666666667,"f1":0.8191881918819188,"true_positive":111,"false_positive":10,"false_negative":39,"true_negative":140,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.5101740402848134,"baseline_adjusted_skill":0.6733333333333333,"baseline_adjusted_score":0.6733333333333333}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · qasper-noul-test-v1 · binary_brier	0.1679 Brier	Setup & source S1MB datasets/qasper / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.16788130620944,"accuracy":0.79,"positive_recall":0.82,"specificity":0.76,"precision":0.7735849056603774,"balanced_accuracy":0.79,"f1":0.7961165048543689,"true_positive":41,"false_positive":12,"false_negative":9,"true_negative":38,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.32847477516224,"baseline_adjusted_skill":0.5800000000000001,"baseline_adjusted_score":0.5800000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · quartz-choice-test-v1 · target_mass_at_prediction	80.00%	Setup & source S1MB datasets/quartz / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.58,"baseline_adjusted_skill":0.523809523809524,"baseline_adjusted_score":0.523809523809524}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · robust-lr-choice-test-v1 · target_mass_at_prediction	51.00%	Setup & source S1MB datasets/robust_lr / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.51,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":0.18333333333333332,"baseline_adjusted_score":0.18333333333333332}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · ruletaker-noul-test-v1 · binary_brier	0.2497 Brier	Setup & source S1MB datasets/ruletaker / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.24970083534727,"accuracy":0.49,"positive_recall":0.06,"specificity":0.92,"precision":0.42857142857142855,"balanced_accuracy":0.49,"f1":0.10526315789473684,"true_positive":3,"false_positive":4,"false_negative":47,"true_negative":46,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.0011966586109199984,"baseline_adjusted_skill":-0.020000000000000018,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · rvlcdip-choice · target_mass_at_prediction	51.00%	Setup & source S1MB datasets/rvlcdip-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.51,"uniform_accuracy_baseline":0.0625,"fixed_answer_accuracy_baseline":0.10666666666666667,"baseline_adjusted_skill":0.4514925373134328,"baseline_adjusted_score":0.4514925373134328}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-contextual-choice-choice-test-v1 · target_mass_at_prediction	97.00%	Setup & source S1MB datasets/s1mb-generalization-contextual-choice / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.97,"uniform_accuracy_baseline":0.20851190476190473,"fixed_answer_accuracy_baseline":0.22333333333333333,"baseline_adjusted_skill":0.9613733905579399,"baseline_adjusted_score":0.9613733905579399}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-contextual-noul-noul-test-v1 · binary_brier	0.0776 Brier	Setup & source S1MB datasets/s1mb-generalization-contextual-noul / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.07758388542651,"accuracy":0.91,"positive_recall":0.96,"specificity":0.86,"precision":0.8727272727272727,"balanced_accuracy":0.9099999999999999,"f1":0.9142857142857143,"true_positive":48,"false_positive":7,"false_negative":2,"true_negative":43,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.68966445829396,"baseline_adjusted_skill":0.8199999999999998,"baseline_adjusted_score":0.8199999999999998}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-contextual-score-score-test-v1 · normalized_expected_score_mae	0.1943 MAE	Setup & source S1MB datasets/s1mb-generalization-contextual-score / test · normalized_expected_score_mae. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.1942518712713709,"normalized_score_rmse":0.2589398364144899,"constant_score_baseline":0.5,"constant_mae_baseline":0.3,"baseline_adjusted_skill":0.3524937624287636,"baseline_adjusted_score":0.3524937624287636}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-diverse-choice-choice-test-v1 · target_mass_at_prediction	87.00%	Setup & source S1MB datasets/s1mb-generalization-diverse-choice / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.87,"uniform_accuracy_baseline":0.22851190476190475,"fixed_answer_accuracy_baseline":0.24333333333333332,"baseline_adjusted_skill":0.8281938325991189,"baseline_adjusted_score":0.8281938325991189}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-diverse-noul-noul-test-v1 · binary_brier	0.1140 Brier	Setup & source S1MB datasets/s1mb-generalization-diverse-noul / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.11397108754695,"accuracy":0.81,"positive_recall":0.88,"specificity":0.74,"precision":0.7719298245614035,"balanced_accuracy":0.81,"f1":0.822429906542056,"true_positive":44,"false_positive":13,"false_negative":6,"true_negative":37,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.5441156498122,"baseline_adjusted_skill":0.6200000000000001,"baseline_adjusted_score":0.6200000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · s1mb-generalization-diverse-score-score-test-v1 · normalized_expected_score_mae	0.1856 MAE	Setup & source S1MB datasets/s1mb-generalization-diverse-score / test · normalized_expected_score_mae. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.18558913143246883,"normalized_score_rmse":0.25990950529454626,"constant_score_baseline":0.4222222222222222,"constant_mae_baseline":0.3042222222222222,"baseline_adjusted_skill":0.38995537513067224,"baseline_adjusted_score":0.38995537513067224}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · scicite-choice-test-v1 · target_mass_at_prediction	62.00%	Setup & source S1MB datasets/scicite / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.62,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.3870967741935484,"baseline_adjusted_score":0.3870967741935484}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · scienceqa-choice · target_mass_at_prediction	84.67%	Setup & source S1MB datasets/scienceqa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.8466666666666667,"uniform_accuracy_baseline":0.3467777777777778,"fixed_answer_accuracy_baseline":0.36,"baseline_adjusted_skill":0.7604166666666667,"baseline_adjusted_score":0.7604166666666667}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · scitail-noul-test-v1 · binary_brier	0.1085 Brier	Setup & source S1MB datasets/scitail / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10847823886465999,"accuracy":0.86,"positive_recall":0.94,"specificity":0.78,"precision":0.8103448275862069,"balanced_accuracy":0.86,"f1":0.8703703703703703,"true_positive":47,"false_positive":11,"false_negative":3,"true_negative":39,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.56608704454136,"baseline_adjusted_skill":0.72,"baseline_adjusted_score":0.72}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · scone-choice-test-v1 · target_mass_at_prediction	59.00%	Setup & source S1MB datasets/scone / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.59,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.5,"baseline_adjusted_skill":0.17999999999999994,"baseline_adjusted_score":0.17999999999999994}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · sdoh-nli-noul-test-v1 · binary_brier	0.1096 Brier	Setup & source S1MB datasets/sdoh_nli / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10960639335402,"accuracy":0.87,"positive_recall":0.82,"specificity":0.92,"precision":0.9111111111111111,"balanced_accuracy":0.87,"f1":0.8631578947368421,"true_positive":41,"false_positive":4,"false_negative":9,"true_negative":46,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.56157442658392,"baseline_adjusted_skill":0.74,"baseline_adjusted_score":0.74}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · sgd-choice-test-v1 · target_mass_at_prediction	58.00%	Setup & source S1MB datasets/sgd / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.58,"uniform_accuracy_baseline":0.3533333333333333,"fixed_answer_accuracy_baseline":0.41,"baseline_adjusted_skill":0.28813559322033894,"baseline_adjusted_score":0.28813559322033894}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · snli-choice-test-v1 · target_mass_at_prediction	82.00%	Setup & source S1MB datasets/snli / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.82,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.7096774193548386,"baseline_adjusted_score":0.7096774193548386}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · sobaco-culture-choice · target_mass_at_prediction	47.67%	Setup & source S1MB datasets/sobaco-culture-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.4766666666666667,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.21500000000000002,"baseline_adjusted_score":0.21500000000000002}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · spartqa-choice-test-v1 · target_mass_at_prediction	39.00%	Setup & source S1MB datasets/spartqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.39,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.01612903225806453,"baseline_adjusted_score":0.01612903225806453}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · stepgame-choice-test-v1 · target_mass_at_prediction	29.00%	Setup & source S1MB datasets/stepgame / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.29,"uniform_accuracy_baseline":0.1111111111111111,"fixed_answer_accuracy_baseline":0.16,"baseline_adjusted_skill":0.15476190476190474,"baseline_adjusted_score":0.15476190476190474}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · sugarcrepe-choice · target_mass_at_prediction	74.00%	Setup & source S1MB datasets/sugarcrepe-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.74,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.58,"baseline_adjusted_skill":0.380952380952381,"baseline_adjusted_score":0.380952380952381}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanoarguana-noul-test-v1 · binary_brier	0.1715 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoArguAna / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.17148534446039002,"accuracy":0.548125,"positive_recall":0.5543836519446276,"specificity":0.5466604966836341,"precision":0.22248677248677248,"balanced_accuracy":0.5505220743141308,"f1":0.3175382291863319,"true_positive":42.05,"false_positive":146.95,"false_negative":33.8,"true_negative":177.2,"positive_prevalence":0.189625,"majority_accuracy_baseline":0.8103750000000001,"constant_brier_baseline":0.07612360937500001,"brier_skill":-1.25272219575952,"baseline_adjusted_skill":0.10104414862826161,"baseline_adjusted_score":0.10104414862826161}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanoarguana-score-test-v1 · normalized_expected_score_mae	0.3220 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoArguAna / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.32196946440553165,"normalized_score_rmse":0.37759765735379563,"constant_score_baseline":0,"constant_mae_baseline":0.21375,"baseline_adjusted_skill":-0.5062898919557037,"baseline_adjusted_score":0}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanodbpedia-noul-test-v1 · binary_brier	0.0829 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoDBPedia / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.08292785515457499,"accuracy":0.761375,"positive_recall":0.516643225503985,"specificity":0.8503494119652293,"precision":0.5565656565656566,"balanced_accuracy":0.6834963187346071,"f1":0.5358619012885972,"true_positive":55.1,"false_positive":43.9,"false_negative":51.55,"true_negative":249.45,"positive_prevalence":0.266625,"majority_accuracy_baseline":0.733375,"constant_brier_baseline":0.115654859375,"brier_skill":0.28297128540281025,"baseline_adjusted_skill":0.3669926374692143,"baseline_adjusted_score":0.3669926374692143}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanodbpedia-score-test-v1 · normalized_expected_score_mae	0.2020 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoDBPedia / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.2020007707526179,"normalized_score_rmse":0.2600190841437415,"constant_score_baseline":0.25,"constant_mae_baseline":0.26875,"baseline_adjusted_skill":0.24836922510653803,"baseline_adjusted_score":0.24836922510653803}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanofiqa2018-noul-test-v1 · binary_brier	0.0712 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoFiQA2018 / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.0711737042120425,"accuracy":0.7721250000000001,"positive_recall":0.43221476510067114,"specificity":0.9039028620988725,"precision":0.6355263157894736,"balanced_accuracy":0.6680588135997718,"f1":0.514513981358189,"true_positive":48.3,"false_positive":27.7,"false_negative":63.45,"true_negative":260.55,"positive_prevalence":0.279375,"majority_accuracy_baseline":0.7206250000000001,"constant_brier_baseline":0.11315585937500001,"brier_skill":0.3710117654961914,"baseline_adjusted_skill":0.33611762719954363,"baseline_adjusted_score":0.33611762719954363}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanofiqa2018-score-test-v1 · normalized_expected_score_mae	0.2063 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoFiQA2018 / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.20626924585016515,"normalized_score_rmse":0.2566367226912141,"constant_score_baseline":0.25,"constant_mae_baseline":0.264375,"baseline_adjusted_skill":0.21978535848637304,"baseline_adjusted_score":0.21978535848637304}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanohotpotqa-noul-test-v1 · binary_brier	0.0888 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoHotpotQA / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.08878304773045499,"accuracy":0.811875,"positive_recall":0.2,"specificity":0.9863453815261044,"precision":0.8068181818181818,"balanced_accuracy":0.5931726907630522,"f1":0.32054176072234764,"true_positive":17.75,"false_positive":4.25,"false_negative":71,"true_negative":307,"positive_prevalence":0.221875,"majority_accuracy_baseline":0.778125,"constant_brier_baseline":0.12674023437499998,"brier_skill":0.299488057850966,"baseline_adjusted_skill":0.18634538152610447,"baseline_adjusted_score":0.18634538152610447}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanohotpotqa-score-test-v1 · normalized_expected_score_mae	0.1914 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoHotpotQA / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.19142745639508843,"normalized_score_rmse":0.2598307307206291,"constant_score_baseline":0,"constant_mae_baseline":0.225625,"baseline_adjusted_skill":0.151568060298777,"baseline_adjusted_score":0.151568060298777}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanomsmarco-noul-test-v1 · binary_brier	0.0552 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoMSMARCO / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.0552197261562225,"accuracy":0.7925,"positive_recall":0.665406427221172,"specificity":0.8382053025152957,"precision":0.5966101694915255,"balanced_accuracy":0.7518058648682339,"f1":0.6291331546023236,"true_positive":70.4,"false_positive":47.6,"false_negative":35.4,"true_negative":246.6,"positive_prevalence":0.2645,"majority_accuracy_baseline":0.7355,"constant_brier_baseline":0.11362725,"brier_skill":0.514027434825515,"baseline_adjusted_skill":0.5036117297364677,"baseline_adjusted_score":0.5036117297364677}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanomsmarco-score-test-v1 · normalized_expected_score_mae	0.1822 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoMSMARCO / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.18219397694980585,"normalized_score_rmse":0.23973324801975154,"constant_score_baseline":0.25,"constant_mae_baseline":0.25125,"baseline_adjusted_skill":0.27484984298584725,"baseline_adjusted_score":0.27484984298584725}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanonfcorpus-noul-test-v1 · binary_brier	0.0640 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoNFCorpus / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.064040439085375,"accuracy":0.82325,"positive_recall":0.4432114882506528,"specificity":0.9132653061224489,"precision":0.5475806451612903,"balanced_accuracy":0.6782383971865509,"f1":0.48989898989898994,"true_positive":33.95,"false_positive":28.05,"false_negative":42.65,"true_negative":295.35,"positive_prevalence":0.19149999999999998,"majority_accuracy_baseline":0.8085,"constant_brier_baseline":0.09190274999999999,"brier_skill":0.30317167782928145,"baseline_adjusted_skill":0.35647679437310176,"baseline_adjusted_score":0.35647679437310176}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanonfcorpus-score-test-v1 · normalized_expected_score_mae	0.1699 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoNFCorpus / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.1699254799839215,"normalized_score_rmse":0.2286537922261499,"constant_score_baseline":0,"constant_mae_baseline":0.206875,"baseline_adjusted_skill":0.1786079517393523,"baseline_adjusted_score":0.1786079517393523}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanonq-noul-test-v1 · binary_brier	0.0599 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoNQ / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.059884814885065,"accuracy":0.8675,"positive_recall":0.5745257452574526,"specificity":0.9337829552421827,"precision":0.6625,"balanced_accuracy":0.7541543502498176,"f1":0.6153846153846153,"true_positive":42.4,"false_positive":21.6,"false_negative":31.4,"true_negative":304.6,"positive_prevalence":0.1845,"majority_accuracy_baseline":0.8155,"constant_brier_baseline":0.10922225,"brier_skill":0.45171597467489455,"baseline_adjusted_skill":0.5083087004996352,"baseline_adjusted_score":0.5083087004996352}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanonq-score-test-v1 · normalized_expected_score_mae	0.2148 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoNQ / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.2148142659726813,"normalized_score_rmse":0.2710634953479237,"constant_score_baseline":0,"constant_mae_baseline":0.189375,"baseline_adjusted_skill":-0.13433275761151853,"baseline_adjusted_score":0}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanoscidocs-noul-test-v1 · binary_brier	0.0921 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoSCIDOCS / test · binary_brier. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.09209576498649,"accuracy":0.81975,"positive_recall":0.1235668789808917,"specificity":0.9897356143079316,"precision":0.7461538461538461,"balanced_accuracy":0.5566512466444117,"f1":0.21202185792349726,"true_positive":9.7,"false_positive":3.3,"false_negative":68.8,"true_negative":318.2,"positive_prevalence":0.19625,"majority_accuracy_baseline":0.80375,"constant_brier_baseline":0.09878593749999999,"brier_skill":0.06772393604616034,"baseline_adjusted_skill":0.11330249328882336,"baseline_adjusted_score":0.11330249328882336}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanoscidocs-score-test-v1 · normalized_expected_score_mae	0.1955 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoSCIDOCS / test · normalized_expected_score_mae. Population 400 cases / 400 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.1954863512896513,"normalized_score_rmse":0.2785469289718142,"constant_score_baseline":0,"constant_mae_baseline":0.21125,"baseline_adjusted_skill":0.07462082229750855,"baseline_adjusted_score":0.07462082229750855}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 400 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanotouche2020-noul-test-v1 · binary_brier	0.1084 Brier	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoTouche2020 / test · binary_brier. Population 392 cases / 392 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.10841369088936735,"accuracy":0.6525510204081633,"positive_recall":0.632858837485172,"specificity":0.6674127126230976,"precision":0.5895027624309392,"balanced_accuracy":0.6501357750541348,"f1":0.6104118993135013,"true_positive":106.7,"false_positive":74.3,"false_negative":61.9,"true_negative":149.1,"positive_prevalence":0.4301020408163265,"majority_accuracy_baseline":0.5698979591836735,"constant_brier_baseline":0.12359641815910039,"brier_skill":0.12284115911991045,"baseline_adjusted_skill":0.30027155010826956,"baseline_adjusted_score":0.30027155010826956}. Counts: {"cases": 392, "expected": 392, "succeeded": 392, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 392 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanobeir--nanotouche2020-score-test-v1 · normalized_expected_score_mae	0.2017 MAE	Setup & source S1MB datasets/synthetic_relevance__nanobeir__NanoTouche2020 / test · normalized_expected_score_mae. Population 392 cases / 392 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.20169652184074194,"normalized_score_rmse":0.25117430565937726,"constant_score_baseline":0.5,"constant_mae_baseline":0.2755102040816326,"baseline_adjusted_skill":0.2679163281336032,"baseline_adjusted_score":0.2679163281336032}. Counts: {"cases": 392, "expected": 392, "succeeded": 392, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 392 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanocoir-noul-test-v1 · binary_brier	0.0707 Brier	Setup & source S1MB datasets/synthetic_relevance__nanocoir / test · binary_brier. Population 800 cases / 800 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.07073697572099125,"accuracy":0.8541250000000001,"positive_recall":0.5169158143194336,"specificity":0.9178183979788974,"precision":0.5429752066115703,"balanced_accuracy":0.7173671061491655,"f1":0.5296251511487304,"true_positive":65.7,"false_positive":55.3,"false_negative":61.4,"true_negative":617.6,"positive_prevalence":0.158875,"majority_accuracy_baseline":0.841125,"constant_brier_baseline":0.09877748437499999,"brier_skill":0.2838755089930761,"baseline_adjusted_skill":0.4347342122983311,"baseline_adjusted_score":0.4347342122983311}. Counts: {"cases": 800, "expected": 800, "succeeded": 800, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 800 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · synthetic-relevance--nanocoir-score-test-v1 · normalized_expected_score_mae	0.2290 MAE	Setup & source S1MB datasets/synthetic_relevance__nanocoir / test · normalized_expected_score_mae. Population 800 cases / 800 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.22902791389372823,"normalized_score_rmse":0.29058960396465044,"constant_score_baseline":0,"constant_mae_baseline":0.165,"baseline_adjusted_skill":-0.38804796299229216,"baseline_adjusted_score":0}. Counts: {"cases": 800, "expected": 800, "succeeded": 800, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 800 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · tallyqa-score · normalized_expected_score_mae	0.0784 MAE	Setup & source S1MB datasets/tallyqa-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"normalized_expected_score_mae":0.07842050456986352,"normalized_score_rmse":0.12694720424385197,"constant_score_baseline":0.1111111111111111,"constant_mae_baseline":0.10629629629629629,"baseline_adjusted_skill":0.2622461242556393,"baseline_adjusted_score":0.2622461242556393}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · temporal-nli-choice-test-v1 · target_mass_at_prediction	49.00%	Setup & source S1MB datasets/temporal_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.49,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.17741935483870966,"baseline_adjusted_score":0.17741935483870966}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · tracie-noul-test-v1 · binary_brier	0.2886 Brier	Setup & source S1MB datasets/tracie / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.28860587595995996,"accuracy":0.54,"positive_recall":0.34,"specificity":0.74,"precision":0.5666666666666667,"balanced_accuracy":0.54,"f1":0.425,"true_positive":17,"false_positive":13,"false_negative":33,"true_negative":37,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.15442350383983983,"baseline_adjusted_skill":0.08000000000000007,"baseline_adjusted_score":0.08000000000000007}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · ud-ewt-choice-test-v1 · target_mass_at_prediction	17.36%	Setup & source S1MB datasets/ud_ewt / test · target_mass_at_prediction. Population 100 cases / 1509 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.17362491716368456,"uniform_accuracy_baseline":0.058823529411764705,"fixed_answer_accuracy_baseline":0.14910536779324055,"baseline_adjusted_skill":0.02881619937694705,"baseline_adjusted_score":0.02881619937694705}. Counts: {"cases": 100, "expected": 1509, "succeeded": 1509, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · visa-noul · binary_brier	0.3050 Brier	Setup & source S1MB datasets/visa-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.3050207093043067,"accuracy":0.4866666666666667,"positive_recall":0.11494252873563218,"specificity":1,"precision":1,"balanced_accuracy":0.5574712643678161,"f1":0.20618556701030927,"true_positive":20,"false_positive":0,"false_negative":154,"true_negative":126,"positive_prevalence":0.58,"majority_accuracy_baseline":0.58,"constant_brier_baseline":0.24359999999999998,"brier_skill":-0.25213755872047083,"baseline_adjusted_skill":0.11494252873563227,"baseline_adjusted_score":0.11494252873563227}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · vsr-noul · binary_brier	0.2723 Brier	Setup & source S1MB datasets/vsr-noul / test · binary_brier. Population 300 cases / 300 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.2723426412957667,"accuracy":0.53,"positive_recall":0.8853503184713376,"specificity":0.13986013986013987,"precision":0.5305343511450382,"balanced_accuracy":0.5126052291657387,"f1":0.6634844868735084,"true_positive":139,"false_positive":123,"false_negative":18,"true_negative":20,"positive_prevalence":0.5233333333333333,"majority_accuracy_baseline":0.5233333333333333,"constant_brier_baseline":0.24945555555555557,"brier_skill":-0.09174815004316073,"baseline_adjusted_skill":0.025210458331477392,"baseline_adjusted_score":0.025210458331477392}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 300 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · winobias-choice-test-v1 · target_mass_at_prediction	55.00%	Setup & source S1MB datasets/winobias / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.55,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":0.02173913043478263,"baseline_adjusted_score":0.02173913043478263}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · winowhy-noul-test-v1 · binary_brier	0.2210 Brier	Setup & source S1MB datasets/winowhy / test · binary_brier. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. All published diagnostics: {"binary_brier":0.22104347334622,"accuracy":0.65,"positive_recall":0.42,"specificity":0.88,"precision":0.7777777777777778,"balanced_accuracy":0.65,"f1":0.5454545454545454,"true_positive":21,"false_positive":6,"false_negative":29,"true_negative":44,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.11582610661512005,"baseline_adjusted_skill":0.30000000000000004,"baseline_adjusted_score":0.30000000000000004}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB · wiqa-choice-test-v1 · target_mass_at_prediction	54.00%	Setup & source S1MB datasets/wiqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions. Exact OmniJev/OneJev1.6-0.8B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-0.8B. Published measured S1MB run; community execution/submission status is not assumed independent. Choice is target mass at the selected option; soft targets are not ordinary hard-label accuracy. All published diagnostics: {"target_mass_at_prediction":0.54,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.37,"baseline_adjusted_skill":0.2698412698412699,"baseline_adjusted_score":0.2698412698412699}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Parameter-count conventions excluding lookup/embedding parameters are not model MoE active counts. No executed weight SHA is stated in this display record; exact named repository and adapter are retained. Pinned public display snapshot 2026-10-17T12:53:13.764Z. Benchmark definitions keep dataset/split and task semantics. Native binary Brier and normalized expected-score MAE are errors, not percents. Failed, partial or missing results never become zero. Checked / reported: 2026-10-18 · Samples: 100 Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · BordaScore	41.26	Setup & source S1MB english-v1 · BordaScore Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · TaskAvg	30.03	Setup & source S1MB english-v1 · TaskAvg Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · Noul	36.63	Setup & source S1MB english-v1 · Noul Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · Choice	39.10	Setup & source S1MB english-v1 · Choice Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · Score	14.37	Setup & source S1MB english-v1 · Score Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · GeneralNoul	72.00	Setup & source S1MB english-v1 · GeneralNoul Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · GeneralChoice	89.48	Setup & source S1MB english-v1 · GeneralChoice Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB english-v1 · GeneralScore	37.12	Setup & source S1MB english-v1 · GeneralScore Published english-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 137 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · BordaScore	19.36	Setup & source S1MB japanese-v1 · BordaScore Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · TaskAvg	27.17	Setup & source S1MB japanese-v1 · TaskAvg Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · Noul	23.76	Setup & source S1MB japanese-v1 · Noul Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · Choice	36.38	Setup & source S1MB japanese-v1 · Choice Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · Score	21.37	Setup & source S1MB japanese-v1 · Score Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · GeneralNoul	36.67	Setup & source S1MB japanese-v1 · GeneralNoul Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · GeneralChoice	56.37	Setup & source S1MB japanese-v1 · GeneralChoice Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB japanese-v1 · GeneralScore	1.96	Setup & source S1MB japanese-v1 · GeneralScore Published japanese-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 27 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · BordaScore	13.82	Setup & source S1MB vision-v1 · BordaScore Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · TaskAvg	30.71	Setup & source S1MB vision-v1 · TaskAvg Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · Noul	29.63	Setup & source S1MB vision-v1 · Noul Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · Choice	51.36	Setup & source S1MB vision-v1 · Choice Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · Score	11.14	Setup & source S1MB vision-v1 · Score Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · GeneralNoul	37.52	Setup & source S1MB vision-v1 · GeneralNoul Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · GeneralChoice	49.37	Setup & source S1MB vision-v1 · GeneralChoice Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
S1MB vision-v1 · GeneralScore	9.16	Setup & source S1MB vision-v1 · GeneralScore Published vision-v1 UI scope; OneJev 1.6 0.8BOmniJev/OneJev1.6-0.8B. Coverage 20 benchmarks. Read the source scoring specification for adjustment, clipping and scope exclusions. Value read from the publisher’s rendered table, not recomputed from rounded components. Noul/Choice/Score and General columns are source baseline-adjusted scores. Borda is relative to the complete model cohort in this dated scope; TaskAvg is equal task weight after equal benchmark weight. Neither is native accuracy/Brier/MAE. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source

Evaluation	Result	Evidence
OneJev vendor · Public index	48.82	Setup & source OneJev vendor · Public index OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor measurements only; explicit source label retained in every view. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Public Knowledge	26.20	Setup & source OneJev vendor · Public Knowledge OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Public Language	48.10	Setup & source OneJev vendor · Public Language OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Public Retrieval	63.40	Setup & source OneJev vendor · Public Retrieval OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Public Tools	73.10	Setup & source OneJev vendor · Public Tools OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Public Arts	35.60	Setup & source OneJev vendor · Public Arts OneJev launch figure, released 1.6 checkpoint. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Vendor-run Decision Index 0.3 public kit; 37 benchmarks / 140,178 requests as reported. Source quotes peers from the 16 October board. Not AA, evaluator-private Full, or S1MB. Exact execution dates, budgets and individual run variances are not published. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · OneJev held-out test set	72.80%	Setup & source OneJev vendor · OneJev held-out test set Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · GUI-agent subset	76.90%	Setup & source OneJev vendor · GUI-agent subset Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Image subset	67.40%	Setup & source OneJev vendor · Image subset Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Long-video subset	62.70%	Setup & source OneJev vendor · Long-video subset Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Short-video subset	56.60%	Setup & source OneJev vendor · Short-video subset Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · Text subset	78.80%	Setup & source OneJev vendor · Text subset Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · DecisionBench medium	57.90%	Setup & source OneJev vendor · DecisionBench medium Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · DecisionBench hard	46.90%	Setup & source OneJev vendor · DecisionBench hard Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · TypeSafe	75.40%	Setup & source OneJev vendor · TypeSafe Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · MMBench	82.50%	Setup & source OneJev vendor · MMBench Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev vendor · MMStar	54.80%	Setup & source OneJev vendor · MMStar Released OneJev 1.6 0.8B; vendor comparison figure. Accuracy as printed. OneJev test set and five subsets are vendor-held-out data; denominators and detailed judge/protocol are not published. DecisionBench is not DecideBench v1.1. Bare TypeSafe/MMBench labels retain this vendor scope rather than assuming a standard revision. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: OneJev 1.6 launch · 2026-10-17 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev H200 screenshot · One question	31.00 ms	Setup & source OneJev H200 screenshot · One question Vendor H200 path; One question. One H200 and one 1280×720 screenshot. Vendor-published timing, not hosted API latency; detailed hardware/runtime precision and distribution statistic unpublished. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: 2c32b28dbd8867471c7b62191bf8c7f1e6803a86 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev H200 screenshot · Ten questions in one request	51.00 ms	Setup & source OneJev H200 screenshot · Ten questions in one request Vendor H200 path; Ten questions in one request. One H200 and one 1280×720 screenshot. Vendor-published timing, not hosted API latency; detailed hardware/runtime precision and distribution statistic unpublished. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: 2c32b28dbd8867471c7b62191bf8c7f1e6803a86 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source
OneJev H200 screenshot · Per question in ten-question request	5.10 ms	Setup & source OneJev H200 screenshot · Per question in ten-question request Vendor H200 path; Per question in ten-question request. One H200 and one 1280×720 screenshot. Vendor-published timing, not hosted API latency; detailed hardware/runtime precision and distribution statistic unpublished. Checked / reported: 2026-10-18 · Samples: Not published Snapshot: 2026-10-17 · Revision: 2c32b28dbd8867471c7b62191bf8c7f1e6803a86 · Published aggregate observations; result reuse terms not separately specified Result source · Benchmark source