Examenos

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

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

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

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

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

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

OneJev vendor · OneJev held-out test set

Released OneJev 1.6 27B; 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

OneJev vendor · GUI-agent subset

Released OneJev 1.6 27B; 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

OneJev vendor · Image subset

Released OneJev 1.6 27B; 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

OneJev vendor · Long-video subset

Released OneJev 1.6 27B; 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

OneJev vendor · Short-video subset

Released OneJev 1.6 27B; 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

OneJev vendor · Text subset

Released OneJev 1.6 27B; 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

OneJev vendor · DecisionBench medium

Released OneJev 1.6 27B; 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

OneJev vendor · DecisionBench hard

Released OneJev 1.6 27B; 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

OneJev vendor · TypeSafe

Released OneJev 1.6 27B; 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

OneJev vendor · MMBench

Released OneJev 1.6 27B; 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

OneJev vendor · MMStar

Released OneJev 1.6 27B; 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

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

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

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

S1MB datasets/aegis2 / test · binary_brier. Population 100 cases / 148 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.1448734445287027,"accuracy":0.7702702702702703,"positive_recall":0.6901408450704225,"specificity":0.8441558441558441,"precision":0.8032786885245902,"balanced_accuracy":0.7671483446131333,"f1":0.7424242424242424,"true_positive":49,"false_positive":12,"false_negative":22,"true_negative":65,"positive_prevalence":0.4797297297297297,"majority_accuracy_baseline":0.5202702702702703,"constant_brier_baseline":0.24958911614317017,"brier_skill":0.4195522354203943,"baseline_adjusted_skill":0.5342966892262666,"baseline_adjusted_score":0.5342966892262666}. Counts: {"cases": 100, "expected": 148, "succeeded": 148, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/aqua_rat / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.67,"uniform_accuracy_baseline":0.2005,"fixed_answer_accuracy_baseline":0.29,"baseline_adjusted_skill":0.5352112676056339,"baseline_adjusted_score":0.5352112676056339}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/arc / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.99,"uniform_accuracy_baseline":0.2533333333333333,"fixed_answer_accuracy_baseline":0.32,"baseline_adjusted_skill":0.9852941176470588,"baseline_adjusted_score":0.9852941176470588}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/arct / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.93,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.53,"baseline_adjusted_skill":0.8510638297872342,"baseline_adjusted_score":0.8510638297872342}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/argument_quality / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.98,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.51,"baseline_adjusted_skill":0.9591836734693877,"baseline_adjusted_score":0.9591836734693877}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/argument_quality / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.13877764465763795,"accuracy":0.5797478472,"positive_recall":0.7027062427959632,"specificity":0.38247379117488745,"precision":0.6461058046716418,"balanced_accuracy":0.5425900169854253,"f1":0.6732184566439771,"true_positive":43.289088913,"false_positive":23.710911087,"false_negative":18.314304193,"true_negative":14.685695807,"positive_prevalence":0.61603393106,"majority_accuracy_baseline":0.61603393106,"constant_brier_baseline":0.07908707655202402,"brier_skill":-0.7547449053367028,"baseline_adjusted_skill":0.08518003397085061,"baseline_adjusted_score":0.08518003397085061}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/babi_nli / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06588602004618,"accuracy":0.92,"positive_recall":0.9,"specificity":0.94,"precision":0.9375,"balanced_accuracy":0.9199999999999999,"f1":0.9183673469387755,"true_positive":45,"false_positive":3,"false_negative":5,"true_negative":47,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.73645591981528,"baseline_adjusted_skill":0.8399999999999999,"baseline_adjusted_score":0.8399999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/banking77 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.012987012987012986,"fixed_answer_accuracy_baseline":0.05,"baseline_adjusted_skill":0.8736842105263158,"baseline_adjusted_score":0.8736842105263158}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/bbq / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.91,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":0.8500000000000001,"baseline_adjusted_score":0.8500000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/boardgameqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.6,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.36,"baseline_adjusted_skill":0.375,"baseline_adjusted_score":0.375}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/canttalk / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.15138849301869,"accuracy":0.79,"positive_recall":1,"specificity":0.58,"precision":0.704225352112676,"balanced_accuracy":0.79,"f1":0.8264462809917356,"true_positive":50,"false_positive":21,"false_negative":0,"true_negative":29,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.39444602792524,"baseline_adjusted_skill":0.5800000000000001,"baseline_adjusted_score":0.5800000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/civil_comments / test · binary_brier. Population 100 cases / 700 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06999208932926612,"accuracy":0.7082978146064228,"positive_recall":0.6097352813318597,"specificity":0.7541921485191769,"precision":0.5359690642875173,"balanced_accuracy":0.6819637149255183,"f1":0.5704774524912056,"true_positive":135.6001732647419,"false_positive":117.3998267352581,"false_negative":86.79170304024592,"true_negative":360.2082969597541,"positive_prevalence":0.3177026804356969,"majority_accuracy_baseline":0.6822973195643032,"constant_brier_baseline":0.08872692397768642,"brier_skill":0.21115163028904105,"baseline_adjusted_skill":0.36392742985103665,"baseline_adjusted_score":0.36392742985103665}. Counts: {"cases": 100, "expected": 700, "succeeded": 700, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/cladder / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.061845432782980006,"accuracy":0.93,"positive_recall":0.88,"specificity":0.98,"precision":0.9777777777777777,"balanced_accuracy":0.9299999999999999,"f1":0.9263157894736842,"true_positive":44,"false_positive":1,"false_negative":6,"true_negative":49,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.75261826886808,"baseline_adjusted_skill":0.8599999999999999,"baseline_adjusted_score":0.8599999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/clinc / test · target_mass_at_prediction. Population 151 cases / 151 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9668874172185431,"uniform_accuracy_baseline":0.006622516556291391,"fixed_answer_accuracy_baseline":0.039735099337748346,"baseline_adjusted_skill":0.9655172413793104,"baseline_adjusted_score":0.9655172413793104}. Counts: {"cases": 151, "expected": 151, "succeeded": 151, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 151

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/commonsensemoralja-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.09304747699899667,"accuracy":0.8666666666666667,"positive_recall":0.7466666666666667,"specificity":0.9866666666666667,"precision":0.9824561403508771,"balanced_accuracy":0.8666666666666667,"f1":0.8484848484848485,"true_positive":112,"false_positive":2,"false_negative":38,"true_negative":148,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6278100920040133,"baseline_adjusted_skill":0.7333333333333334,"baseline_adjusted_score":0.7333333333333334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/contract_nli / test · target_mass_at_prediction. Population 100 cases / 1700 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8905882352941177,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.4576470588235294,"baseline_adjusted_skill":0.7982646420824295,"baseline_adjusted_score":0.7982646420824295}. Counts: {"cases": 100, "expected": 1700, "succeeded": 1700, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/corr2cause / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.49,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":0.14999999999999997,"baseline_adjusted_score":0.14999999999999997}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/creak / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07467344212305,"accuracy":0.93,"positive_recall":0.96,"specificity":0.9,"precision":0.9056603773584906,"balanced_accuracy":0.9299999999999999,"f1":0.9320388349514563,"true_positive":48,"false_positive":5,"false_negative":2,"true_negative":45,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.7013062315078,"baseline_adjusted_skill":0.8599999999999999,"baseline_adjusted_score":0.8599999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/crows_pairs / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.52,"baseline_adjusted_skill":0.5833333333333334,"baseline_adjusted_score":0.5833333333333334}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/cvbench-depth-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9633333333333334,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5,"baseline_adjusted_skill":0.9266666666666667,"baseline_adjusted_score":0.9266666666666667}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/cvbench-distance-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.87,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.52,"baseline_adjusted_skill":0.7291666666666666,"baseline_adjusted_score":0.7291666666666666}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/cvbench-relation-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9266666666666666,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5266666666666666,"baseline_adjusted_skill":0.8450704225352113,"baseline_adjusted_score":0.8450704225352113}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/dbpedia / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.97,"uniform_accuracy_baseline":0.07142857142857142,"fixed_answer_accuracy_baseline":0.11,"baseline_adjusted_skill":0.9662921348314606,"baseline_adjusted_score":0.9662921348314606}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/defeasible_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":0.7391304347826086,"baseline_adjusted_score":0.7391304347826086}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/esci / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.5,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.27,"baseline_adjusted_skill":0.3150684931506849,"baseline_adjusted_score":0.3150684931506849}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/ethics / test · target_mass_at_prediction. Population 18 cases / 18 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7777777777777778,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.6666666666666666,"baseline_adjusted_skill":0.3333333333333334,"baseline_adjusted_score":0.3333333333333334}. Counts: {"cases": 18, "expected": 18, "succeeded": 18, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 18

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/ethics / test · binary_brier. Population 82 cases / 82 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.11446552897593902,"accuracy":0.8292682926829268,"positive_recall":0.85,"specificity":0.8095238095238095,"precision":0.8095238095238095,"balanced_accuracy":0.8297619047619047,"f1":0.8292682926829268,"true_positive":34,"false_positive":8,"false_negative":6,"true_negative":34,"positive_prevalence":0.4878048780487805,"majority_accuracy_baseline":0.5121951219512195,"constant_brier_baseline":0.2498512790005949,"brier_skill":0.5418653471224917,"baseline_adjusted_skill":0.6595238095238094,"baseline_adjusted_score":0.6595238095238094}. Counts: {"cases": 82, "expected": 82, "succeeded": 82, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 82

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/ethos / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.11976098804599002,"accuracy":0.84,"positive_recall":0.8,"specificity":0.88,"precision":0.8695652173913043,"balanced_accuracy":0.8400000000000001,"f1":0.8333333333333334,"true_positive":40,"false_positive":6,"false_negative":10,"true_negative":44,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.52095604781604,"baseline_adjusted_skill":0.6800000000000002,"baseline_adjusted_score":0.6800000000000002}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/few_nerd / test · target_mass_at_prediction. Population 100 cases / 358 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8379888268156425,"uniform_accuracy_baseline":0.125,"fixed_answer_accuracy_baseline":0.2206703910614525,"baseline_adjusted_skill":0.7921146953405018,"baseline_adjusted_score":0.7921146953405018}. Counts: {"cases": 100, "expected": 358, "succeeded": 358, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/fol_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.58,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.39,"baseline_adjusted_skill":0.3114754098360655,"baseline_adjusted_score":0.3114754098360655}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/followir_core17 / test · binary_brier. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.160661264517345,"accuracy":0.795,"positive_recall":0.4090909090909091,"specificity":0.9038461538461539,"precision":0.5454545454545454,"balanced_accuracy":0.6564685314685315,"f1":0.4675324675324675,"true_positive":18,"false_positive":15,"false_negative":26,"true_negative":141,"positive_prevalence":0.22,"majority_accuracy_baseline":0.78,"constant_brier_baseline":0.1716,"brier_skill":0.06374554477071681,"baseline_adjusted_skill":0.3129370629370629,"baseline_adjusted_score":0.3129370629370629}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/followir_news21 / test · binary_brier. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.173553159134175,"accuracy":0.755,"positive_recall":0.18,"specificity":0.9466666666666667,"precision":0.5294117647058824,"balanced_accuracy":0.5633333333333334,"f1":0.26865671641791045,"true_positive":9,"false_positive":8,"false_negative":41,"true_negative":142,"positive_prevalence":0.25,"majority_accuracy_baseline":0.75,"constant_brier_baseline":0.1875,"brier_skill":0.0743831512844001,"baseline_adjusted_skill":0.1266666666666667,"baseline_adjusted_score":0.1266666666666667}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/followir_robust04 / test · binary_brier. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.042316709757445,"accuracy":0.935,"positive_recall":0.5263157894736842,"specificity":0.9779005524861878,"precision":0.7142857142857143,"balanced_accuracy":0.752108170979936,"f1":0.6060606060606061,"true_positive":10,"false_positive":4,"false_negative":9,"true_negative":177,"positive_prevalence":0.095,"majority_accuracy_baseline":0.905,"constant_brier_baseline":0.085975,"brier_skill":0.5078021546095377,"baseline_adjusted_skill":0.504216341959872,"baseline_adjusted_score":0.504216341959872}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-diverse-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9933333333333333,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.9899999999999998,"baseline_adjusted_score":0.9899999999999998}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-diverse-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.004980816109740001,"accuracy":0.9933333333333333,"positive_recall":1,"specificity":0.9866666666666667,"precision":0.9868421052631579,"balanced_accuracy":0.9933333333333334,"f1":0.9933774834437086,"true_positive":150,"false_positive":2,"false_negative":0,"true_negative":148,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.98007673556104,"baseline_adjusted_skill":0.9866666666666668,"baseline_adjusted_score":0.9866666666666668}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-knowledge-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9933333333333333,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.991111111111111,"baseline_adjusted_score":0.991111111111111}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-knowledge-hard-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07927707125157,"accuracy":0.8866666666666667,"positive_recall":0.9933333333333333,"specificity":0.78,"precision":0.8186813186813187,"balanced_accuracy":0.8866666666666667,"f1":0.8975903614457831,"true_positive":149,"false_positive":33,"false_negative":1,"true_negative":117,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6828917149937199,"baseline_adjusted_skill":0.7733333333333334,"baseline_adjusted_score":0.7733333333333334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-knowledge-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.1212228720726785,"normalized_score_rmse":0.1696008051462463,"constant_score_baseline":0.5,"constant_mae_baseline":0.3333333333333333,"baseline_adjusted_skill":0.6363313837819644,"baseline_adjusted_score":0.6363313837819644}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-math-formal-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.4066666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.2088888888888889,"baseline_adjusted_score":0.2088888888888889}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-math-formal-hard-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.2543648269167,"accuracy":0.5533333333333333,"positive_recall":0.2866666666666667,"specificity":0.82,"precision":0.6142857142857143,"balanced_accuracy":0.5533333333333333,"f1":0.39090909090909093,"true_positive":43,"false_positive":27,"false_negative":107,"true_negative":123,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.017459307666799928,"baseline_adjusted_skill":0.10666666666666669,"baseline_adjusted_score":0.10666666666666669}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-japanese-math-formal-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.2886449515932655,"normalized_score_rmse":0.3383504659492212,"constant_score_baseline":0.5,"constant_mae_baseline":0.3,"baseline_adjusted_skill":0.037850161355781675,"baseline_adjusted_score":0.037850161355781675}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-easy-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9766666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3,"baseline_adjusted_skill":0.9666666666666669,"baseline_adjusted_score":0.9666666666666669}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-easy-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.008578450195063333,"accuracy":0.9933333333333333,"positive_recall":0.9865771812080537,"specificity":1,"precision":1,"balanced_accuracy":0.9932885906040269,"f1":0.9932432432432432,"true_positive":147,"false_positive":0,"false_negative":2,"true_negative":151,"positive_prevalence":0.49666666666666665,"majority_accuracy_baseline":0.5033333333333334,"constant_brier_baseline":0.2499888888888889,"brier_skill":0.9656846740941508,"baseline_adjusted_skill":0.9865771812080537,"baseline_adjusted_score":0.9865771812080537}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-easy-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.13203267838574242,"normalized_score_rmse":0.18051895749329336,"constant_score_baseline":0.5,"constant_mae_baseline":0.36666666666666664,"baseline_adjusted_skill":0.6399108771297934,"baseline_adjusted_score":0.6399108771297934}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-hard-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8166666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.2866666666666667,"baseline_adjusted_skill":0.7429906542056075,"baseline_adjusted_score":0.7429906542056075}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-hard-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.09985279635588,"accuracy":0.86,"positive_recall":0.8535031847133758,"specificity":0.8671328671328671,"precision":0.8758169934640523,"balanced_accuracy":0.8603180259231215,"f1":0.864516129032258,"true_positive":134,"false_positive":19,"false_negative":23,"true_negative":124,"positive_prevalence":0.5233333333333333,"majority_accuracy_baseline":0.5233333333333333,"constant_brier_baseline":0.24945555555555557,"brier_skill":0.5997170873444746,"baseline_adjusted_skill":0.7206360518462429,"baseline_adjusted_score":0.7206360518462429}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/generalization-vision-hard-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.20315662753202798,"normalized_score_rmse":0.24940018261133504,"constant_score_baseline":0.5,"constant_mae_baseline":0.3016666666666667,"baseline_adjusted_skill":0.32655261591592943,"baseline_adjusted_score":0.32655261591592943}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/go_emotions / test · binary_brier. Population 100 cases / 2800 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07186412756588247,"accuracy":0.873625,"positive_recall":0.580795555996429,"specificity":0.8923182137678177,"precision":0.25612423447069116,"balanced_accuracy":0.7365568848821233,"f1":0.3554840472359673,"true_positive":97.58333333333333,"false_positive":283.4166666666667,"false_negative":70.43333333333334,"true_negative":2348.5666666666666,"positive_prevalence":0.060005952380952375,"majority_accuracy_baseline":0.9399940476190476,"constant_brier_baseline":0.03220097218679138,"brier_skill":-1.2317378229766813,"baseline_adjusted_skill":0.4731137697642467,"baseline_adjusted_score":0.4731137697642467}. Counts: {"cases": 100, "expected": 2800, "succeeded": 2800, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/gpqa-ja-main-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.49666666666666665,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.32888888888888884,"baseline_adjusted_score":0.32888888888888884}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/gretel_pii / test · target_mass_at_prediction. Population 100 cases / 353 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7960339943342776,"uniform_accuracy_baseline":0.027777777777777776,"fixed_answer_accuracy_baseline":0.1274787535410765,"baseline_adjusted_skill":0.7662337662337662,"baseline_adjusted_score":0.7662337662337662}. Counts: {"cases": 100, "expected": 353, "succeeded": 353, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/gsm8k / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3,"baseline_adjusted_skill":0.8571428571428573,"baseline_adjusted_score":0.8571428571428573}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/hans / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07152434497737001,"accuracy":0.9,"positive_recall":1,"specificity":0.8,"precision":0.8333333333333334,"balanced_accuracy":0.9,"f1":0.9090909090909091,"true_positive":50,"false_positive":10,"false_negative":0,"true_negative":40,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.71390262009052,"baseline_adjusted_skill":0.8,"baseline_adjusted_score":0.8}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/hatecheck / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.08182845624537,"accuracy":0.87,"positive_recall":0.98,"specificity":0.76,"precision":0.8032786885245902,"balanced_accuracy":0.87,"f1":0.8828828828828829,"true_positive":49,"false_positive":12,"false_negative":1,"true_negative":38,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.67268617501852,"baseline_adjusted_skill":0.74,"baseline_adjusted_score":0.74}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/hh_rlhf / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.65,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.61,"baseline_adjusted_skill":0.10256410256410266,"baseline_adjusted_score":0.10256410256410266}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/hwu64 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.88,"uniform_accuracy_baseline":0.015625,"fixed_answer_accuracy_baseline":0.06,"baseline_adjusted_skill":0.872340425531915,"baseline_adjusted_score":0.872340425531915}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/impli / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06181147323283,"accuracy":0.91,"positive_recall":0.88,"specificity":0.94,"precision":0.9361702127659575,"balanced_accuracy":0.9099999999999999,"f1":0.9072164948453608,"true_positive":44,"false_positive":3,"false_negative":6,"true_negative":47,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.7527541070686801,"baseline_adjusted_skill":0.8199999999999998,"baseline_adjusted_score":0.8199999999999998}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jamc-qa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.53,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.37333333333333335,"baseline_adjusted_score":0.37333333333333335}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jamp-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.76,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.64,"baseline_adjusted_score":0.64}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/janli-choice / test · target_mass_at_prediction. Population 285 cases / 285 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8631578947368421,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5263157894736842,"baseline_adjusted_skill":0.7111111111111111,"baseline_adjusted_score":0.7111111111111111}. Counts: {"cases": 285, "expected": 285, "succeeded": 285, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 285

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jcola-noul / test · binary_brier. Population 299 cases / 299 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.193732608100398,"accuracy":0.6889632107023411,"positive_recall":0.9,"specificity":0.47651006711409394,"precision":0.6338028169014085,"balanced_accuracy":0.688255033557047,"f1":0.743801652892562,"true_positive":135,"false_positive":78,"false_negative":15,"true_negative":71,"positive_prevalence":0.5016722408026756,"majority_accuracy_baseline":0.5016722408026756,"constant_brier_baseline":0.24999720361069788,"brier_skill":0.22506089947276597,"baseline_adjusted_skill":0.37651006711409396,"baseline_adjusted_score":0.37651006711409396}. Counts: {"cases": 299, "expected": 299, "succeeded": 299, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 299

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jcommonsenseqa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9633333333333334,"uniform_accuracy_baseline":0.2,"fixed_answer_accuracy_baseline":0.2,"baseline_adjusted_skill":0.9541666666666667,"baseline_adjusted_score":0.9541666666666667}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jculture-mcq-noul / test · binary_brier. Population 283 cases / 283 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.2528789888873074,"accuracy":0.6007067137809188,"positive_recall":0.8120300751879699,"specificity":0.41333333333333333,"precision":0.5510204081632653,"balanced_accuracy":0.6126817042606516,"f1":0.6565349544072948,"true_positive":108,"false_positive":88,"false_negative":25,"true_negative":62,"positive_prevalence":0.46996466431095407,"majority_accuracy_baseline":0.5300353356890459,"constant_brier_baseline":0.24909787861004634,"brier_skill":-0.015179215087496978,"baseline_adjusted_skill":0.2253634085213032,"baseline_adjusted_score":0.2253634085213032}. Counts: {"cases": 283, "expected": 283, "succeeded": 283, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 283

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jmmlu-japan-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8666666666666667,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.8222222222222223,"baseline_adjusted_score":0.8222222222222223}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jmteb-retrieval / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9866666666666667,"uniform_accuracy_baseline":0.20616666666666666,"fixed_answer_accuracy_baseline":0.21333333333333335,"baseline_adjusted_skill":0.9830508474576272,"baseline_adjusted_score":0.9830508474576272}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jnli-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8666666666666667,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.8,"baseline_adjusted_score":0.8}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jsem-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7533333333333333,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.37333333333333335,"baseline_adjusted_skill":0.6063829787234042,"baseline_adjusted_score":0.6063829787234042}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jsick-sts-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.130295911903346,"normalized_score_rmse":0.1595307076979285,"constant_score_baseline":0.5,"constant_mae_baseline":0.26208333333333333,"baseline_adjusted_skill":0.5028454871732426,"baseline_adjusted_score":0.5028454871732426}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/jsts-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.10057150064243091,"normalized_score_rmse":0.12998726306306532,"constant_score_baseline":0.52,"constant_mae_baseline":0.24733333333333335,"baseline_adjusted_skill":0.5933766820386891,"baseline_adjusted_score":0.5933766820386891}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/koniq-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.13717111070966242,"normalized_score_rmse":0.1554849505948164,"constant_score_baseline":0.578319964349225,"constant_mae_baseline":0.11212962028602311,"baseline_adjusted_skill":-0.2233262750713223,"baseline_adjusted_score":0}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/kuci-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8433333333333334,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.7911111111111112,"baseline_adjusted_score":0.7911111111111112}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/laya__ag_news / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.87,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.33,"baseline_adjusted_skill":0.8059701492537314,"baseline_adjusted_score":0.8059701492537314}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/laya__enron_spam / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.03664524350343,"accuracy":0.96,"positive_recall":0.94,"specificity":0.98,"precision":0.9791666666666666,"balanced_accuracy":0.96,"f1":0.9591836734693877,"true_positive":47,"false_positive":1,"false_negative":3,"true_negative":49,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.8534190259862799,"baseline_adjusted_skill":0.9199999999999999,"baseline_adjusted_score":0.9199999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/laya__typed_decisions / test · target_mass_at_prediction. Population 100 cases / 148 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.49379849324324326,"uniform_accuracy_baseline":0.23445945945945948,"fixed_answer_accuracy_baseline":0.2623439054054054,"baseline_adjusted_skill":0.3137703186266522,"baseline_adjusted_score":0.3137703186266522}. Counts: {"cases": 100, "expected": 148, "succeeded": 148, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/laya__typed_decisions / test · binary_brier. Population 100 cases / 152 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.02405441714336842,"accuracy":0.7246491118421053,"positive_recall":0.7640050135956967,"specificity":0.6871735558890402,"precision":0.6993004197530864,"balanced_accuracy":0.7255892847423684,"f1":0.7302221594001129,"true_positive":56.643334,"false_positive":24.356666,"false_negative":17.496669,"true_negative":53.503331,"positive_prevalence":0.4877631776315789,"majority_accuracy_baseline":0.512236822368421,"constant_brier_baseline":0.09451910039768556,"brier_skill":0.7455073414562732,"baseline_adjusted_skill":0.4511785694847368,"baseline_adjusted_score":0.4511785694847368}. Counts: {"cases": 100, "expected": 152, "succeeded": 152, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/laya__typed_decisions / test · normalized_expected_score_mae. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.09818261170135663,"normalized_score_rmse":0.1514394985444621,"constant_score_baseline":0.5841668333333333,"constant_mae_baseline":0.21700552708333334,"baseline_adjusted_skill":0.5475570921119763,"baseline_adjusted_score":0.5475570921119763}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/lexcomp / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.18514792020649,"accuracy":0.7,"positive_recall":0.88,"specificity":0.52,"precision":0.6470588235294118,"balanced_accuracy":0.7,"f1":0.7457627118644068,"true_positive":44,"false_positive":24,"false_negative":6,"true_negative":26,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.25940831917403995,"baseline_adjusted_skill":0.3999999999999999,"baseline_adjusted_score":0.3999999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/logical_entailment / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.22944289328183,"accuracy":0.64,"positive_recall":0.4,"specificity":0.88,"precision":0.7692307692307693,"balanced_accuracy":0.64,"f1":0.5263157894736842,"true_positive":20,"false_positive":6,"false_negative":30,"true_negative":44,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.08222842687267995,"baseline_adjusted_skill":0.28,"baseline_adjusted_score":0.28}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/lonli / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.94,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.39,"baseline_adjusted_skill":0.901639344262295,"baseline_adjusted_score":0.901639344262295}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/massive / test · target_mass_at_prediction. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.915,"uniform_accuracy_baseline":0.036111111111111115,"fixed_answer_accuracy_baseline":0.08833333333333333,"baseline_adjusted_skill":0.906764168190128,"baseline_adjusted_score":0.906764168190128}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/mintaka-ja-boolean-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.16570905491103,"accuracy":0.7233333333333334,"positive_recall":0.7333333333333333,"specificity":0.7133333333333334,"precision":0.7189542483660131,"balanced_accuracy":0.7233333333333334,"f1":0.7260726072607261,"true_positive":110,"false_positive":43,"false_negative":40,"true_negative":107,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.33716378035588,"baseline_adjusted_skill":0.44666666666666677,"baseline_adjusted_score":0.44666666666666677}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/miqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.96,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.56,"baseline_adjusted_skill":0.909090909090909,"baseline_adjusted_score":0.909090909090909}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/mmlu-prox-ja-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.6433333333333333,"uniform_accuracy_baseline":0.11021428571428572,"fixed_answer_accuracy_baseline":0.12554761904761905,"baseline_adjusted_skill":0.5921256841016146,"baseline_adjusted_score":0.5921256841016146}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/mmmlu-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7633333333333333,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.25,"baseline_adjusted_skill":0.6844444444444444,"baseline_adjusted_score":0.6844444444444444}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/mtop / test · target_mass_at_prediction. Population 100 cases / 200 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.93,"uniform_accuracy_baseline":0.049879324215607396,"fixed_answer_accuracy_baseline":0.12442477876106194,"baseline_adjusted_skill":0.9200525571053164,"baseline_adjusted_score":0.9200525571053164}. Counts: {"cases": 100, "expected": 200, "succeeded": 200, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/mvtec-ad-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.13321758873365333,"accuracy":0.7866666666666666,"positive_recall":0.7314814814814815,"specificity":0.9285714285714286,"precision":0.9634146341463414,"balanced_accuracy":0.8300264550264551,"f1":0.8315789473684211,"true_positive":158,"false_positive":6,"false_negative":58,"true_negative":78,"positive_prevalence":0.72,"majority_accuracy_baseline":0.72,"constant_brier_baseline":0.2016,"brier_skill":0.3391984685830688,"baseline_adjusted_skill":0.6600529100529102,"baseline_adjusted_score":0.6600529100529102}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/nights-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9333333333333333,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.5033333333333333,"baseline_adjusted_skill":0.865771812080537,"baseline_adjusted_score":0.865771812080537}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/nlsat / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.32159634363080997,"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":-0.28638537452323987,"baseline_adjusted_skill":0,"baseline_adjusted_score":0}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/nlvr2-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06755472857186334,"accuracy":0.9166666666666666,"positive_recall":0.89375,"specificity":0.9428571428571428,"precision":0.9470198675496688,"balanced_accuracy":0.9183035714285714,"f1":0.9196141479099679,"true_positive":143,"false_positive":8,"false_negative":17,"true_negative":132,"positive_prevalence":0.5333333333333333,"majority_accuracy_baseline":0.5333333333333333,"constant_brier_baseline":0.2488888888888889,"brier_skill":0.7285747512737634,"baseline_adjusted_skill":0.8366071428571429,"baseline_adjusted_score":0.8366071428571429}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__amount-extraction-control-v1 / test · target_mass_at_prediction. Population 36 cases / 36 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8888888888888888,"uniform_accuracy_baseline":0.23888888888888887,"fixed_answer_accuracy_baseline":0.2916666666666667,"baseline_adjusted_skill":0.8431372549019607,"baseline_adjusted_score":0.8431372549019607}. Counts: {"cases": 36, "expected": 36, "succeeded": 36, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 36

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__amount-extraction-control-v1 / test · binary_brier. Population 64 cases / 64 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.04438273374489063,"accuracy":0.9375,"positive_recall":0.90625,"specificity":0.96875,"precision":0.9666666666666667,"balanced_accuracy":0.9375,"f1":0.9354838709677419,"true_positive":29,"false_positive":1,"false_negative":3,"true_negative":31,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.8224690650204375,"baseline_adjusted_skill":0.875,"baseline_adjusted_score":0.875}. Counts: {"cases": 64, "expected": 64, "succeeded": 64, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 64

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__browser-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":1,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__citation-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.99,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.34,"baseline_adjusted_skill":0.9848484848484849,"baseline_adjusted_score":0.9848484848484849}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__context-retention-control-v1 / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.030300173454149995,"accuracy":0.94,"positive_recall":0.88,"specificity":1,"precision":1,"balanced_accuracy":0.94,"f1":0.9361702127659575,"true_positive":44,"false_positive":0,"false_negative":6,"true_negative":50,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.8787993061834001,"baseline_adjusted_skill":0.8799999999999999,"baseline_adjusted_score":0.8799999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__customer-control-v1 / test · target_mass_at_prediction. Population 17 cases / 17 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8529411764705882,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.3235294117647059,"baseline_adjusted_skill":0.7826086956521737,"baseline_adjusted_score":0.7826086956521737}. Counts: {"cases": 17, "expected": 17, "succeeded": 17, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 17

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__customer-control-v1 / test · binary_brier. Population 33 cases / 33 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.04357199136275758,"accuracy":0.9393939393939394,"positive_recall":1,"specificity":0.8823529411764706,"precision":0.8888888888888888,"balanced_accuracy":0.9411764705882353,"f1":0.9411764705882353,"true_positive":16,"false_positive":2,"false_negative":0,"true_negative":15,"positive_prevalence":0.48484848484848486,"majority_accuracy_baseline":0.5151515151515151,"constant_brier_baseline":0.2497704315886134,"brier_skill":0.8255518434042537,"baseline_adjusted_skill":0.8823529411764706,"baseline_adjusted_score":0.8823529411764706}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 33

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__customer-control-v1 / test · normalized_expected_score_mae. Population 50 cases / 50 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.12169322467962393,"normalized_score_rmse":0.2390579759341982,"constant_score_baseline":0.25,"constant_mae_baseline":0.33,"baseline_adjusted_skill":0.6312326524859881,"baseline_adjusted_score":0.6312326524859881}. Counts: {"cases": 50, "expected": 50, "succeeded": 50, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 50

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__drone-control-v1 / test · target_mass_at_prediction. Population 32 cases / 32 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.1875,"uniform_accuracy_baseline":0.178125,"fixed_answer_accuracy_baseline":0.2875,"baseline_adjusted_skill":-0.14035087719298242,"baseline_adjusted_score":0}. Counts: {"cases": 32, "expected": 32, "succeeded": 32, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 32

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__drone-control-v1 / test · binary_brier. Population 34 cases / 34 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.12613190844744118,"accuracy":0.8823529411764706,"positive_recall":0.7647058823529411,"specificity":1,"precision":1,"balanced_accuracy":0.8823529411764706,"f1":0.8666666666666667,"true_positive":13,"false_positive":0,"false_negative":4,"true_negative":17,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.49547236621023527,"baseline_adjusted_skill":0.7647058823529411,"baseline_adjusted_score":0.7647058823529411}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 34

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__drone-control-v1 / test · normalized_expected_score_mae. Population 34 cases / 34 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.39480259269619644,"normalized_score_rmse":0.420816340000293,"constant_score_baseline":0.5,"constant_mae_baseline":0.14705882352941177,"baseline_adjusted_skill":-1.6846576303341356,"baseline_adjusted_score":0}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 34

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__email-selection-control-v1 / test · target_mass_at_prediction. Population 48 cases / 48 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9375,"uniform_accuracy_baseline":0.13194444444444445,"fixed_answer_accuracy_baseline":0.6006944444444444,"baseline_adjusted_skill":0.8434782608695652,"baseline_adjusted_score":0.8434782608695652}. Counts: {"cases": 48, "expected": 48, "succeeded": 48, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 48

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__email-selection-control-v1 / test · binary_brier. Population 52 cases / 52 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.0002663551107500002,"accuracy":1,"positive_recall":1,"specificity":1,"precision":1,"balanced_accuracy":1,"f1":1,"true_positive":26,"false_positive":0,"false_negative":0,"true_negative":26,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.998934579557,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 52, "expected": 52, "succeeded": 52, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 52

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__entity-alignment-control-v1 / test · binary_brier. Population 75 cases / 75 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.0010690367868266667,"accuracy":1,"positive_recall":1,"specificity":1,"precision":1,"balanced_accuracy":1,"f1":1,"true_positive":39,"false_positive":0,"false_negative":0,"true_negative":36,"positive_prevalence":0.52,"majority_accuracy_baseline":0.52,"constant_brier_baseline":0.24960000000000004,"brier_skill":0.9957170000527777,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 75, "expected": 75, "succeeded": 75, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 75

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__entity-alignment-control-v1 / test · normalized_expected_score_mae. Population 25 cases / 25 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.12883316351716378,"normalized_score_rmse":0.18220989022724843,"constant_score_baseline":0.5,"constant_mae_baseline":0.28,"baseline_adjusted_skill":0.5398815588672723,"baseline_adjusted_score":0.5398815588672723}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 25

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__ir-control-v1 / test · target_mass_at_prediction. Population 17 cases / 17 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8823529411764706,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.2647058823529412,"baseline_adjusted_skill":0.84,"baseline_adjusted_score":0.84}. Counts: {"cases": 17, "expected": 17, "succeeded": 17, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 17

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__ir-control-v1 / test · binary_brier. Population 28 cases / 28 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06837132209757144,"accuracy":0.9285714285714286,"positive_recall":1,"specificity":0.8571428571428571,"precision":0.875,"balanced_accuracy":0.9285714285714286,"f1":0.9333333333333333,"true_positive":14,"false_positive":2,"false_negative":0,"true_negative":12,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.7265147116097143,"baseline_adjusted_skill":0.8571428571428572,"baseline_adjusted_score":0.8571428571428572}. Counts: {"cases": 28, "expected": 28, "succeeded": 28, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 28

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__ir-control-v1 / test · normalized_expected_score_mae. Population 55 cases / 55 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.057291089707083735,"normalized_score_rmse":0.15029041739720822,"constant_score_baseline":0.3333333333333333,"constant_mae_baseline":0.1696969696969697,"baseline_adjusted_skill":0.6623917927975422,"baseline_adjusted_score":0.6623917927975422}. Counts: {"cases": 55, "expected": 55, "succeeded": 55, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 55

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__mailroom-control-v1 / test · target_mass_at_prediction. Population 15 cases / 15 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":1,"uniform_accuracy_baseline":0.15555555555555553,"fixed_answer_accuracy_baseline":0.2,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 15, "expected": 15, "succeeded": 15, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 15

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__mailroom-control-v1 / test · binary_brier. Population 85 cases / 85 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06042046298268236,"accuracy":0.8941176470588236,"positive_recall":0.8,"specificity":1,"precision":1,"balanced_accuracy":0.9,"f1":0.8888888888888888,"true_positive":36,"false_positive":0,"false_negative":9,"true_negative":40,"positive_prevalence":0.5294117647058824,"majority_accuracy_baseline":0.5294117647058824,"constant_brier_baseline":0.2491349480968858,"brier_skill":0.7574789749722889,"baseline_adjusted_skill":0.8,"baseline_adjusted_score":0.8}. Counts: {"cases": 85, "expected": 85, "succeeded": 85, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 85

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__painting-geometry-v1 / test · target_mass_at_prediction. Population 25 cases / 25 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":1,"uniform_accuracy_baseline":0.07805555555555556,"fixed_answer_accuracy_baseline":0.24,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 25

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__painting-geometry-v1 / test · binary_brier. Population 50 cases / 50 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.01639666406876,"accuracy":1,"positive_recall":1,"specificity":1,"precision":1,"balanced_accuracy":1,"f1":1,"true_positive":35,"false_positive":0,"false_negative":0,"true_negative":15,"positive_prevalence":0.7,"majority_accuracy_baseline":0.7,"constant_brier_baseline":0.21,"brier_skill":0.921920647291619,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 50, "expected": 50, "succeeded": 50, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 50

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__painting-geometry-v1 / test · normalized_expected_score_mae. Population 25 cases / 25 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.017911558164232975,"normalized_score_rmse":0.028255082525588598,"constant_score_baseline":0.5,"constant_mae_baseline":0.28,"baseline_adjusted_skill":0.9360301494134536,"baseline_adjusted_score":0.9360301494134536}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 25

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__phone-extraction-control-v1 / test · target_mass_at_prediction. Population 79 cases / 79 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8354430379746836,"uniform_accuracy_baseline":0.24240506329113923,"fixed_answer_accuracy_baseline":0.4177215189873418,"baseline_adjusted_skill":0.717391304347826,"baseline_adjusted_score":0.717391304347826}. Counts: {"cases": 79, "expected": 79, "succeeded": 79, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 79

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__phone-extraction-control-v1 / test · binary_brier. Population 21 cases / 21 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":6.372524366666653e-05,"accuracy":1,"positive_recall":1,"specificity":1,"precision":1,"balanced_accuracy":1,"f1":1,"true_positive":11,"false_positive":0,"false_negative":0,"true_negative":10,"positive_prevalence":0.5238095238095238,"majority_accuracy_baseline":0.5238095238095238,"constant_brier_baseline":0.2494331065759637,"brier_skill":0.9997445197049364,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 21, "expected": 21, "succeeded": 21, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 21

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__reasoning-control-v1 / test · target_mass_at_prediction. Population 33 cases / 33 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9393939393939394,"uniform_accuracy_baseline":0.32373737373737377,"fixed_answer_accuracy_baseline":0.3939393939393939,"baseline_adjusted_skill":0.9000000000000001,"baseline_adjusted_score":0.9000000000000001}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 33

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__reasoning-control-v1 / test · binary_brier. Population 33 cases / 33 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.16371968312236362,"accuracy":0.7272727272727273,"positive_recall":0.6470588235294118,"specificity":0.8125,"precision":0.7857142857142857,"balanced_accuracy":0.7297794117647058,"f1":0.7096774193548387,"true_positive":11,"false_positive":3,"false_negative":6,"true_negative":13,"positive_prevalence":0.5151515151515151,"majority_accuracy_baseline":0.5151515151515151,"constant_brier_baseline":0.2497704315886134,"brier_skill":0.3445193569108309,"baseline_adjusted_skill":0.4595588235294117,"baseline_adjusted_score":0.4595588235294117}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 33

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__reasoning-control-v1 / test · normalized_expected_score_mae. Population 34 cases / 34 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.04674572538637588,"normalized_score_rmse":0.09255773687660876,"constant_score_baseline":0.7083333333333333,"constant_mae_baseline":0.3431372549019608,"baseline_adjusted_skill":0.8637696003025617,"baseline_adjusted_score":0.8637696003025617}. Counts: {"cases": 34, "expected": 34, "succeeded": 34, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 34

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__silent-failure-control-v1 / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.15737593892564,"accuracy":0.82,"positive_recall":0.78,"specificity":0.86,"precision":0.8478260869565217,"balanced_accuracy":0.8200000000000001,"f1":0.8125,"true_positive":39,"false_positive":7,"false_negative":11,"true_negative":43,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.37049624429744,"baseline_adjusted_skill":0.6400000000000001,"baseline_adjusted_score":0.6400000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__snake-v1 / test · target_mass_at_prediction. Population 25 cases / 25 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.8933333333333333,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.35333333333333333,"baseline_adjusted_skill":0.8350515463917526,"baseline_adjusted_score":0.8350515463917526}. Counts: {"cases": 25, "expected": 25, "succeeded": 25, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 25

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__snake-v1 / test · binary_brier. Population 75 cases / 75 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.10295595791024,"accuracy":0.8666666666666667,"positive_recall":1,"specificity":0.7368421052631579,"precision":0.7872340425531915,"balanced_accuracy":0.868421052631579,"f1":0.8809523809523809,"true_positive":37,"false_positive":10,"false_negative":0,"true_negative":28,"positive_prevalence":0.49333333333333335,"majority_accuracy_baseline":0.5066666666666666,"constant_brier_baseline":0.24995555555555554,"brier_skill":0.5881029422154338,"baseline_adjusted_skill":0.736842105263158,"baseline_adjusted_score":0.736842105263158}. Counts: {"cases": 75, "expected": 75, "succeeded": 75, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 75

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__sponsor-segment-control-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.96,"uniform_accuracy_baseline":0.14285714285714285,"fixed_answer_accuracy_baseline":0.22,"baseline_adjusted_skill":0.9487179487179487,"baseline_adjusted_score":0.9487179487179487}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__tic_tac_toe-v1 / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.43,"uniform_accuracy_baseline":0.32416666666666666,"fixed_answer_accuracy_baseline":0.34,"baseline_adjusted_skill":0.13636363636363633,"baseline_adjusted_score":0.13636363636363633}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__vizdoom-basic-v1 / test · binary_brier. Population 33 cases / 33 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.02706482320615151,"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":0.8916412041489008,"baseline_adjusted_skill":0.9411764705882353,"baseline_adjusted_score":0.9411764705882353}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 33

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__vizdoom-basic-v1 / test · normalized_expected_score_mae. Population 33 cases / 33 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.03293564152061537,"normalized_score_rmse":0.05378159429753592,"constant_score_baseline":0.5,"constant_mae_baseline":0.2196969696969697,"baseline_adjusted_skill":0.8500860454923714,"baseline_adjusted_score":0.8500860454923714}. Counts: {"cases": 33, "expected": 33, "succeeded": 33, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 33

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__workflow-controls-v1__agent_trace_observability / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07585499724325001,"accuracy":0.88,"positive_recall":0.7272727272727273,"specificity":0.9230769230769231,"precision":0.7272727272727273,"balanced_accuracy":0.8251748251748252,"f1":0.7272727272727273,"true_positive":16,"false_positive":6,"false_negative":6,"true_negative":72,"positive_prevalence":0.22,"majority_accuracy_baseline":0.78,"constant_brier_baseline":0.1716,"brier_skill":0.557954561519522,"baseline_adjusted_skill":0.6503496503496504,"baseline_adjusted_score":0.6503496503496504}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__workflow-controls-v1__customer_service / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.1209973819589,"accuracy":0.81,"positive_recall":0.4864864864864865,"specificity":1,"precision":1,"balanced_accuracy":0.7432432432432432,"f1":0.6545454545454545,"true_positive":18,"false_positive":0,"false_negative":19,"true_negative":63,"positive_prevalence":0.37,"majority_accuracy_baseline":0.63,"constant_brier_baseline":0.23310000000000003,"brier_skill":0.4809207123170314,"baseline_adjusted_skill":0.4864864864864864,"baseline_adjusted_score":0.4864864864864864}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__workflow-controls-v1__invoice_processing / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.04739289314407,"accuracy":0.95,"positive_recall":0.8974358974358975,"specificity":0.9836065573770492,"precision":0.9722222222222222,"balanced_accuracy":0.9405212274064734,"f1":0.9333333333333333,"true_positive":35,"false_positive":1,"false_negative":4,"true_negative":60,"positive_prevalence":0.39,"majority_accuracy_baseline":0.61,"constant_brier_baseline":0.2379,"brier_skill":0.8007864937197562,"baseline_adjusted_skill":0.8810424548129467,"baseline_adjusted_score":0.8810424548129467}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/open_jev__workflow-controls-v1__security_incidents / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.04101538690912,"accuracy":0.93,"positive_recall":0.8387096774193549,"specificity":0.9710144927536232,"precision":0.9285714285714286,"balanced_accuracy":0.904862085086489,"f1":0.8813559322033898,"true_positive":26,"false_positive":2,"false_negative":5,"true_negative":67,"positive_prevalence":0.31,"majority_accuracy_baseline":0.69,"constant_brier_baseline":0.21389999999999998,"brier_skill":0.808249710569799,"baseline_adjusted_skill":0.809724170172978,"baseline_adjusted_score":0.809724170172978}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/openbookqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.95,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.28,"baseline_adjusted_skill":0.9305555555555555,"baseline_adjusted_score":0.9305555555555555}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/patent_similarity / test · normalized_expected_score_mae. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.14798238342193,"normalized_score_rmse":0.182924720357707,"constant_score_baseline":0.25,"constant_mae_baseline":0.2325,"baseline_adjusted_skill":0.36351663044331184,"baseline_adjusted_score":0.36351663044331184}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/paws / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.08540149558693999,"accuracy":0.89,"positive_recall":0.92,"specificity":0.86,"precision":0.8679245283018868,"balanced_accuracy":0.89,"f1":0.8932038834951457,"true_positive":46,"false_positive":7,"false_negative":4,"true_negative":43,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.65839401765224,"baseline_adjusted_skill":0.78,"baseline_adjusted_score":0.78}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/plane / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.32529743850741,"accuracy":0.51,"positive_recall":0.78,"specificity":0.24,"precision":0.5064935064935064,"balanced_accuracy":0.51,"f1":0.6141732283464567,"true_positive":39,"false_positive":38,"false_negative":11,"true_negative":12,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.30118975402964,"baseline_adjusted_skill":0.020000000000000018,"baseline_adjusted_score":0.020000000000000018}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/poem_sentiment / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.79,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.65,"baseline_adjusted_skill":0.4000000000000001,"baseline_adjusted_score":0.4000000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/pope-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.09846466835737666,"accuracy":0.88,"positive_recall":0.7933333333333333,"specificity":0.9666666666666667,"precision":0.9596774193548387,"balanced_accuracy":0.88,"f1":0.8686131386861314,"true_positive":119,"false_positive":5,"false_negative":31,"true_negative":145,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6061413265704934,"baseline_adjusted_skill":0.76,"baseline_adjusted_score":0.76}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/qasper / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07618679541554,"accuracy":0.89,"positive_recall":0.88,"specificity":0.9,"precision":0.8979591836734694,"balanced_accuracy":0.89,"f1":0.8888888888888888,"true_positive":44,"false_positive":5,"false_negative":6,"true_negative":45,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6952528183378399,"baseline_adjusted_skill":0.78,"baseline_adjusted_score":0.78}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/quartz / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.95,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.58,"baseline_adjusted_skill":0.8809523809523808,"baseline_adjusted_score":0.8809523809523808}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/robust_lr / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.69,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.4,"baseline_adjusted_skill":0.4833333333333332,"baseline_adjusted_score":0.4833333333333332}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/ruletaker / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.08648424168757,"accuracy":0.84,"positive_recall":0.74,"specificity":0.94,"precision":0.925,"balanced_accuracy":0.84,"f1":0.8222222222222222,"true_positive":37,"false_positive":3,"false_negative":13,"true_negative":47,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.6540630332497199,"baseline_adjusted_skill":0.6799999999999999,"baseline_adjusted_score":0.6799999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/rvlcdip-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7533333333333333,"uniform_accuracy_baseline":0.0625,"fixed_answer_accuracy_baseline":0.10666666666666667,"baseline_adjusted_skill":0.7238805970149254,"baseline_adjusted_score":0.7238805970149254}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-contextual-choice / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":1,"uniform_accuracy_baseline":0.20851190476190473,"fixed_answer_accuracy_baseline":0.22333333333333333,"baseline_adjusted_skill":1,"baseline_adjusted_score":1}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-contextual-noul / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.01212217318242,"accuracy":0.99,"positive_recall":0.98,"specificity":1,"precision":1,"balanced_accuracy":0.99,"f1":0.98989898989899,"true_positive":49,"false_positive":0,"false_negative":1,"true_negative":50,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.95151130727032,"baseline_adjusted_skill":0.98,"baseline_adjusted_score":0.98}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-contextual-score / test · normalized_expected_score_mae. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.05160318556036567,"normalized_score_rmse":0.11584328177116793,"constant_score_baseline":0.5,"constant_mae_baseline":0.3,"baseline_adjusted_skill":0.8279893814654478,"baseline_adjusted_score":0.8279893814654478}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-diverse-choice / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.98,"uniform_accuracy_baseline":0.22851190476190475,"fixed_answer_accuracy_baseline":0.24333333333333332,"baseline_adjusted_skill":0.973568281938326,"baseline_adjusted_score":0.973568281938326}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-diverse-noul / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.00854899708927,"accuracy":0.99,"positive_recall":1,"specificity":0.98,"precision":0.9803921568627451,"balanced_accuracy":0.99,"f1":0.9900990099009901,"true_positive":50,"false_positive":1,"false_negative":0,"true_negative":49,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.96580401164292,"baseline_adjusted_skill":0.98,"baseline_adjusted_score":0.98}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/s1mb-generalization-diverse-score / test · normalized_expected_score_mae. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.029746404985261706,"normalized_score_rmse":0.05571225219168503,"constant_score_baseline":0.4222222222222222,"constant_mae_baseline":0.3042222222222222,"baseline_adjusted_skill":0.9022214591426752,"baseline_adjusted_score":0.9022214591426752}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/scicite / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.79,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.6612903225806452,"baseline_adjusted_score":0.6612903225806452}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/scienceqa-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.98,"uniform_accuracy_baseline":0.3467777777777778,"fixed_answer_accuracy_baseline":0.36,"baseline_adjusted_skill":0.96875,"baseline_adjusted_score":0.96875}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/scitail / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.07293538348544999,"accuracy":0.92,"positive_recall":0.96,"specificity":0.88,"precision":0.8888888888888888,"balanced_accuracy":0.9199999999999999,"f1":0.9230769230769231,"true_positive":48,"false_positive":6,"false_negative":2,"true_negative":44,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.7082584660582001,"baseline_adjusted_skill":0.8399999999999999,"baseline_adjusted_score":0.8399999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/scone / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.93,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.5,"baseline_adjusted_skill":0.8600000000000001,"baseline_adjusted_score":0.8600000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/sdoh_nli / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.043295899932599996,"accuracy":0.95,"positive_recall":0.94,"specificity":0.96,"precision":0.9591836734693877,"balanced_accuracy":0.95,"f1":0.9494949494949495,"true_positive":47,"false_positive":2,"false_negative":3,"true_negative":48,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.8268164002696,"baseline_adjusted_skill":0.8999999999999999,"baseline_adjusted_score":0.8999999999999999}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/sgd / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.6,"uniform_accuracy_baseline":0.3533333333333333,"fixed_answer_accuracy_baseline":0.41,"baseline_adjusted_skill":0.3220338983050847,"baseline_adjusted_score":0.3220338983050847}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/snli / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.95,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.9193548387096774,"baseline_adjusted_score":0.9193548387096774}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/sobaco-culture-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.7733333333333333,"uniform_accuracy_baseline":0.3333333333333333,"fixed_answer_accuracy_baseline":0.3333333333333333,"baseline_adjusted_skill":0.6599999999999999,"baseline_adjusted_score":0.6599999999999999}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/spartqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.59,"uniform_accuracy_baseline":0.25,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.33870967741935476,"baseline_adjusted_score":0.33870967741935476}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/stepgame / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.52,"uniform_accuracy_baseline":0.1111111111111111,"fixed_answer_accuracy_baseline":0.16,"baseline_adjusted_skill":0.42857142857142855,"baseline_adjusted_score":0.42857142857142855}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/sugarcrepe-choice / test · target_mass_at_prediction. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.9433333333333334,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.58,"baseline_adjusted_skill":0.8650793650793651,"baseline_adjusted_score":0.8650793650793651}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoArguAna / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.0494481785969725,"accuracy":0.835375,"positive_recall":0.34937376400791037,"specificity":0.94909763998149,"precision":0.6162790697674418,"balanced_accuracy":0.6492357019947002,"f1":0.44594026083298277,"true_positive":26.5,"false_positive":16.5,"false_negative":49.35,"true_negative":307.65,"positive_prevalence":0.189625,"majority_accuracy_baseline":0.8103750000000001,"constant_brier_baseline":0.07612360937500001,"brier_skill":0.3504225692533711,"baseline_adjusted_skill":0.2984714039894003,"baseline_adjusted_score":0.2984714039894003}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoArguAna / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.17590620124056536,"normalized_score_rmse":0.2099436948146631,"constant_score_baseline":0,"constant_mae_baseline":0.21375,"baseline_adjusted_skill":0.1770470117400451,"baseline_adjusted_score":0.1770470117400451}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoDBPedia / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.06065683404205,"accuracy":0.800875,"positive_recall":0.5157055789967182,"specificity":0.9045508777910346,"precision":0.6626506024096386,"balanced_accuracy":0.7101282283938763,"f1":0.5800158186132349,"true_positive":55,"false_positive":28,"false_negative":51.65,"true_negative":265.35,"positive_prevalence":0.266625,"majority_accuracy_baseline":0.733375,"constant_brier_baseline":0.115654859375,"brier_skill":0.4755357935685528,"baseline_adjusted_skill":0.4202564567877527,"baseline_adjusted_score":0.4202564567877527}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoDBPedia / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.17814243016093717,"normalized_score_rmse":0.22854140843083207,"constant_score_baseline":0.25,"constant_mae_baseline":0.26875,"baseline_adjusted_skill":0.3371444459127919,"baseline_adjusted_score":0.3371444459127919}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoFiQA2018 / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.0513129720009625,"accuracy":0.7983750000000001,"positive_recall":0.6894854586129754,"specificity":0.8405897658282742,"precision":0.6264227642276422,"balanced_accuracy":0.7650376122206248,"f1":0.6564430244941427,"true_positive":77.05,"false_positive":45.95,"false_negative":34.7,"true_negative":242.3,"positive_prevalence":0.279375,"majority_accuracy_baseline":0.7206250000000001,"constant_brier_baseline":0.11315585937500001,"brier_skill":0.5465283699458229,"baseline_adjusted_skill":0.5300752244412497,"baseline_adjusted_score":0.5300752244412497}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoFiQA2018 / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.17035504228224232,"normalized_score_rmse":0.21481790028102132,"constant_score_baseline":0.25,"constant_mae_baseline":0.264375,"baseline_adjusted_skill":0.3556310457409274,"baseline_adjusted_score":0.3556310457409274}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoHotpotQA / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.0369634646215425,"accuracy":0.896875,"positive_recall":0.7014084507042253,"specificity":0.9526104417670683,"precision":0.8084415584415584,"balanced_accuracy":0.8270094462356468,"f1":0.751131221719457,"true_positive":62.25,"false_positive":14.75,"false_negative":26.5,"true_negative":296.5,"positive_prevalence":0.221875,"majority_accuracy_baseline":0.778125,"constant_brier_baseline":0.12674023437499998,"brier_skill":0.7083525622007711,"baseline_adjusted_skill":0.6540188924712935,"baseline_adjusted_score":0.6540188924712935}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoHotpotQA / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.1356668813205936,"normalized_score_rmse":0.18169712103356464,"constant_score_baseline":0,"constant_mae_baseline":0.225625,"baseline_adjusted_skill":0.3987063431774245,"baseline_adjusted_score":0.3987063431774245}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoMSMARCO / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.036519426353805,"accuracy":0.8402499999999999,"positive_recall":0.7414933837429111,"specificity":0.8757647858599592,"precision":0.6821739130434783,"balanced_accuracy":0.8086290848014351,"f1":0.7105978260869567,"true_positive":78.45,"false_positive":36.55,"false_negative":27.35,"true_negative":257.65,"positive_prevalence":0.2645,"majority_accuracy_baseline":0.7355,"constant_brier_baseline":0.11362725,"brier_skill":0.6786032720689359,"baseline_adjusted_skill":0.6172581696028703,"baseline_adjusted_score":0.6172581696028703}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoMSMARCO / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.14946490859392034,"normalized_score_rmse":0.19321522743780678,"constant_score_baseline":0.25,"constant_mae_baseline":0.25125,"baseline_adjusted_skill":0.4051147916659886,"baseline_adjusted_score":0.4051147916659886}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoNFCorpus / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.055202667322232495,"accuracy":0.823,"positive_recall":0.52088772845953,"specificity":0.8945578231292517,"precision":0.5391891891891891,"balanced_accuracy":0.7077227757943909,"f1":0.5298804780876493,"true_positive":39.9,"false_positive":34.1,"false_negative":36.7,"true_negative":289.3,"positive_prevalence":0.19149999999999998,"majority_accuracy_baseline":0.8085,"constant_brier_baseline":0.09190274999999999,"brier_skill":0.39933606641550445,"baseline_adjusted_skill":0.4154455515887818,"baseline_adjusted_score":0.4154455515887818}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoNFCorpus / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.1578255887211749,"normalized_score_rmse":0.21318155540378042,"constant_score_baseline":0,"constant_mae_baseline":0.206875,"baseline_adjusted_skill":0.23709685210308218,"baseline_adjusted_score":0.23709685210308218}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoNQ / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.03834447709528,"accuracy":0.88925,"positive_recall":0.6944444444444444,"specificity":0.9333231146535867,"precision":0.702054794520548,"balanced_accuracy":0.8138837795490156,"f1":0.6982288828337874,"true_positive":51.25,"false_positive":21.75,"false_negative":22.55,"true_negative":304.45,"positive_prevalence":0.1845,"majority_accuracy_baseline":0.8155,"constant_brier_baseline":0.10922225,"brier_skill":0.648931631647581,"baseline_adjusted_skill":0.6277675590980312,"baseline_adjusted_score":0.6277675590980312}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoNQ / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.1493747721664909,"normalized_score_rmse":0.21125559067123068,"constant_score_baseline":0,"constant_mae_baseline":0.189375,"baseline_adjusted_skill":0.2112223251934474,"baseline_adjusted_score":0.2112223251934474}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoSCIDOCS / test · binary_brier. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.057336946808685,"accuracy":0.83975,"positive_recall":0.2254777070063694,"specificity":0.9897356143079316,"precision":0.8428571428571429,"balanced_accuracy":0.6076066606571505,"f1":0.3557788944723618,"true_positive":17.7,"false_positive":3.3,"false_negative":60.8,"true_negative":318.2,"positive_prevalence":0.19625,"majority_accuracy_baseline":0.80375,"constant_brier_baseline":0.09878593749999999,"brier_skill":0.4195839179166062,"baseline_adjusted_skill":0.21521332131430104,"baseline_adjusted_score":0.21521332131430104}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoSCIDOCS / test · normalized_expected_score_mae. Population 400 cases / 400 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.18019077618191062,"normalized_score_rmse":0.2331935770466183,"constant_score_baseline":0,"constant_mae_baseline":0.21125,"baseline_adjusted_skill":0.14702591156492006,"baseline_adjusted_score":0.14702591156492006}. Counts: {"cases": 400, "expected": 400, "succeeded": 400, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 400

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoTouche2020 / test · binary_brier. Population 392 cases / 392 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.08164657146140561,"accuracy":0.6948979591836734,"positive_recall":0.8659549228944247,"specificity":0.5658012533572068,"precision":0.6008230452674898,"balanced_accuracy":0.7158780881258158,"f1":0.7094266277939747,"true_positive":146,"false_positive":97,"false_negative":22.6,"true_negative":126.4,"positive_prevalence":0.4301020408163265,"majority_accuracy_baseline":0.5698979591836735,"constant_brier_baseline":0.12359641815910039,"brier_skill":0.339409890047902,"baseline_adjusted_skill":0.43175617625163154,"baseline_adjusted_score":0.43175617625163154}. Counts: {"cases": 392, "expected": 392, "succeeded": 392, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 392

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanobeir__NanoTouche2020 / test · normalized_expected_score_mae. Population 392 cases / 392 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.19427361206908864,"normalized_score_rmse":0.24459497676184747,"constant_score_baseline":0.5,"constant_mae_baseline":0.2755102040816326,"baseline_adjusted_skill":0.29485874137886336,"baseline_adjusted_score":0.29485874137886336}. Counts: {"cases": 392, "expected": 392, "succeeded": 392, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 392

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanocoir / test · binary_brier. Population 800 cases / 800 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.03708384072087875,"accuracy":0.90375,"positive_recall":0.6022816679779701,"specificity":0.9606925248922573,"precision":0.7432038834951457,"balanced_accuracy":0.7814870964351137,"f1":0.6653628857018689,"true_positive":76.55,"false_positive":26.45,"false_negative":50.55,"true_negative":646.45,"positive_prevalence":0.158875,"majority_accuracy_baseline":0.841125,"constant_brier_baseline":0.09877748437499999,"brier_skill":0.6245719259250091,"baseline_adjusted_skill":0.5629741928702274,"baseline_adjusted_score":0.5629741928702274}. Counts: {"cases": 800, "expected": 800, "succeeded": 800, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 800

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/synthetic_relevance__nanocoir / test · normalized_expected_score_mae. Population 800 cases / 800 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.13850452273451658,"normalized_score_rmse":0.17990192514427888,"constant_score_baseline":0,"constant_mae_baseline":0.165,"baseline_adjusted_skill":0.16057865009383898,"baseline_adjusted_score":0.16057865009383898}. Counts: {"cases": 800, "expected": 800, "succeeded": 800, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 800

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/tallyqa-score / test · normalized_expected_score_mae. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"normalized_expected_score_mae":0.04223300132413792,"normalized_score_rmse":0.08128521646708257,"constant_score_baseline":0.1111111111111111,"constant_mae_baseline":0.10629629629629629,"baseline_adjusted_skill":0.6026860502607234,"baseline_adjusted_score":0.6026860502607234}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/temporal_nli / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.59,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.38,"baseline_adjusted_skill":0.33870967741935476,"baseline_adjusted_score":0.33870967741935476}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/tracie / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.27206470883028,"accuracy":0.6,"positive_recall":0.34,"specificity":0.86,"precision":0.7083333333333334,"balanced_accuracy":0.6,"f1":0.4594594594594595,"true_positive":17,"false_positive":7,"false_negative":33,"true_negative":43,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":-0.08825883532112,"baseline_adjusted_skill":0.19999999999999996,"baseline_adjusted_score":0.19999999999999996}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/ud_ewt / test · target_mass_at_prediction. Population 100 cases / 1509 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.5964214711729622,"uniform_accuracy_baseline":0.058823529411764705,"fixed_answer_accuracy_baseline":0.14910536779324055,"baseline_adjusted_skill":0.5257009345794392,"baseline_adjusted_score":0.5257009345794392}. Counts: {"cases": 100, "expected": 1509, "succeeded": 1509, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/visa-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.21533922722174,"accuracy":0.6333333333333333,"positive_recall":0.43103448275862066,"specificity":0.9126984126984127,"precision":0.872093023255814,"balanced_accuracy":0.6718664477285167,"f1":0.5769230769230769,"true_positive":75,"false_positive":11,"false_negative":99,"true_negative":115,"positive_prevalence":0.58,"majority_accuracy_baseline":0.58,"constant_brier_baseline":0.24359999999999998,"brier_skill":0.11601302454129714,"baseline_adjusted_skill":0.34373289545703334,"baseline_adjusted_score":0.34373289545703334}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/vsr-noul / test · binary_brier. Population 300 cases / 300 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.14716189838116667,"accuracy":0.7866666666666666,"positive_recall":0.89171974522293,"specificity":0.6713286713286714,"precision":0.7486631016042781,"balanced_accuracy":0.7815242082758007,"f1":0.813953488372093,"true_positive":140,"false_positive":47,"false_negative":17,"true_negative":96,"positive_prevalence":0.5233333333333333,"majority_accuracy_baseline":0.5233333333333333,"constant_brier_baseline":0.24945555555555557,"brier_skill":0.4100676649456595,"baseline_adjusted_skill":0.5630484165516014,"baseline_adjusted_score":0.5630484165516014}. Counts: {"cases": 300, "expected": 300, "succeeded": 300, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 300

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/winobias / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.94,"uniform_accuracy_baseline":0.5,"fixed_answer_accuracy_baseline":0.54,"baseline_adjusted_skill":0.8695652173913042,"baseline_adjusted_score":0.8695652173913042}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/winowhy / test · binary_brier. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Native binary Brier/normalized expected-score MAE is an error, not a percent. Diagnostics: {"binary_brier":0.1735333081445,"accuracy":0.8,"positive_recall":0.88,"specificity":0.72,"precision":0.7586206896551724,"balanced_accuracy":0.8,"f1":0.8148148148148148,"true_positive":44,"false_positive":14,"false_negative":6,"true_negative":36,"positive_prevalence":0.5,"majority_accuracy_baseline":0.5,"constant_brier_baseline":0.25,"brier_skill":0.30586676742200003,"baseline_adjusted_skill":0.6000000000000001,"baseline_adjusted_score":0.6000000000000001}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified

S1MB datasets/wiqa / test · target_mass_at_prediction. Population 100 cases / 100 decisions.

Exact OmniJev/OneJev1.6-27B; adapter onejev; settings {"questions_per_call": 1}. Run OmniJev__OneJev1.6-27B.

Published complete measured S1MB summary, checksum verified. Execution/submission status is not assumed independent. Choice target mass is a fraction converted to percent; soft targets are not ordinary hard-label accuracy. Diagnostics: {"target_mass_at_prediction":0.58,"uniform_accuracy_baseline":0.33333333333333326,"fixed_answer_accuracy_baseline":0.37,"baseline_adjusted_skill":0.33333333333333326,"baseline_adjusted_score":0.33333333333333326}. Counts: {"cases": 100, "expected": 100, "succeeded": 100, "failed": 0, "abstained": 0}. Exact executed weight SHA is not published. Source non-embedding parameter counts do not replace official tensor totals and are not MoE active counts.

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: 2026-10-19 · Samples: 100

Snapshot: 2026-10-17T12:53:13.764Z · Revision: 706ef6637478b78e908be90088f68d8c8f0d52e4 · Published aggregate observations; result reuse terms not separately specified