Gemma 4 E2B Benchmarks, Specifications & Availability

Gemma 4 E2B

Explore Gemma 4 E2B from Google: published specifications, source-linked vendor benchmarks, independent evaluator coverage and recorded pricing when available.

Published specifications

Provider: Google
Access: Open weights
License: Apache-2.0
Context window: 128K
Total parameters: 5.1B
Active parameters: Not published
Released: 2026-03-31
Modalities: text, image, video, audio
Family: Gemma 4

Hugging Face · Announcement · Website

Model notes

Public ungated Apache-2.0 instruction-tuned checkpoint; PT, assistant/MTP and QAT builds are not extra curated models. Vendor architecture size is retained separately from HF tensor totals including modality modules. Effective dense size is 2.3B; the larger total includes per-layer lookup embeddings. Effective is not a MoE active count. Audio input is supported for up to30 seconds; video uses frames (up to60 seconds at1fps), text output only. Context is the published configured limit, not a universal guarantee of whole-window accuracy. Reasoning is an on/off mode, not an API effort level. Report Table9 disables thinking; vision uses1120 or280 visual-token budgets. OmniDocBench uses native average edit distance, not accuracy. Card/report audio aggregates and the older launch τ² figures differ and remain source-specific observations. Official release notes date31March; evaluator release dates may say2April and do not override the vendor date. No exact public OpenRouter route found19October.

Compare Gemma 4 E2B with other models →Compare subscription plans Explore data coverage

Pricing · OpenRouter

No OpenRouter pricing is recorded for this model. Missing rates are not free.

Official / vendor benchmarks

Default headline records. Own-vendor, peer-vendor and third-party provenance remain visible in evidence; configurations may differ.

Official / vendor headline scores; expand evidence for every record
Benchmark / evaluator	Headline score	Evidence
MMLU-Pro	60%unknown effort	All 1 recorded result & sources 60% · raw 60 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
AIME 2026	37.5%unknown effort	All 1 recorded result & sources 37.5% · raw 37.5 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. No tools. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
LiveCodeBench v6	44%unknown effort	All 1 recorded result & sources 44% · raw 44 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
Codeforces Rating	633unknown effort	All 1 recorded result & sources 633 · raw 633 rating Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
SciCode (Gemma report; aggregation unspecified)	21%unknown effort	All 1 recorded result & sources 21% · raw 21 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
GPQA Diamond	43.4%unknown effort	All 1 recorded result & sources 43.4% · raw 43.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
BigBench Extra Hard (micro average)	21.9%unknown effort	All 1 recorded result & sources 21.9% · raw 21.9 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
IFBench (Gemma report protocol)	38%unknown effort	All 1 recorded result & sources 38% · raw 38 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. IFBench strict/loose convention is not stated in this table. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
IFEval	94.6%unknown effort	All 1 recorded result & sources 94.6% · raw 94.6 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
MMMLU	67.4%unknown effort	All 1 recorded result & sources 67.4% · raw 67.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
MRCR v2 (8 needles,128K average)	19.1%unknown effort	All 1 recorded result & sources 19.1% · raw 19.1 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
Terminal Bench Hard (Gemma report)	3%unknown effort	All 1 recorded result & sources 3% · raw 3 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
τ²-bench (Airline)	31%unknown effort	All 1 recorded result & sources 31% · raw 31 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
τ²-bench (Retail)	29.4%unknown effort	All 2 recorded results & sources 34.6% · raw 34.6 % Alternative · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. Demoted19October: own launch figure has a different retail result; retain this report observation without averaging. https://arxiv.org/pdf/2607.02770#page=5 29.4% · raw 29.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google launch benchmark figure; τ² Retail printed29.4. Own-launch tiebreak over later report values; source disagreement is retained explicitly. https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/
τ²-bench (Telecom)	19.7%unknown effort	All 1 recorded result & sources 19.7% · raw 19.7 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google final instruction-tuned report Table5; thinking enabled, no effort tier inferred. Report dated19June2026, not execution date. Visually checked numeric table. https://arxiv.org/pdf/2607.02770#page=5
MMMU Pro (no tools)	44.2%unknown effort	All 1 recorded result & sources 44.2% · raw 44.2 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table6; thinking,1120 visual tokens; execution date and exact output budget unpublished. Native %. https://arxiv.org/pdf/2607.02770#page=6
MMMU Pro (Gemma,280 visual tokens)	43.2%unknown effort	All 1 recorded result & sources 43.2% · raw 43.2 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table12; thinking,280 visual tokens; separate visual-resolution configuration, not a new checkpoint. https://arxiv.org/pdf/2607.02770#page=17
MATH-Vision	52.4%unknown effort	All 1 recorded result & sources 52.4% · raw 52.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table6; thinking,1120 visual tokens; execution date and exact output budget unpublished. Native %. https://arxiv.org/pdf/2607.02770#page=6
MATH-Vision (Gemma,280 visual tokens)	53%unknown effort	All 1 recorded result & sources 53% · raw 53 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table12; thinking,280 visual tokens; separate visual-resolution configuration, not a new checkpoint. https://arxiv.org/pdf/2607.02770#page=17
MedXpertQA MM	23.5%unknown effort	All 1 recorded result & sources 23.5% · raw 23.5 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table6; thinking,1120 visual tokens; execution date and exact output budget unpublished. Native %. https://arxiv.org/pdf/2607.02770#page=6
MedXpertQA MM (Gemma,280 visual tokens)	22.5%unknown effort	All 1 recorded result & sources 22.5% · raw 22.5 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table12; thinking,280 visual tokens; separate visual-resolution configuration, not a new checkpoint. https://arxiv.org/pdf/2607.02770#page=17
InfographicVQA	63.9%unknown effort	All 1 recorded result & sources 63.9% · raw 63.9 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table6; thinking,1120 visual tokens; execution date and exact output budget unpublished. Native %. https://arxiv.org/pdf/2607.02770#page=6
InfographicVQA (Gemma,280 visual tokens)	44.6%unknown effort	All 1 recorded result & sources 44.6% · raw 44.6 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table12; thinking,280 visual tokens; separate visual-resolution configuration, not a new checkpoint. https://arxiv.org/pdf/2607.02770#page=17
OmniDocBench1.5 (average edit distance)	0.290 edit distanceunknown effort	All 1 recorded result & sources 0.290 edit distance · raw 0.29 edit distance Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table6; thinking,1120 visual tokens; execution date and exact output budget unpublished. Native edit distance. https://arxiv.org/pdf/2607.02770#page=6
OmniDocBench1.5 (average edit distance) (Gemma,280 visual tokens)	0.496 edit distanceunknown effort	All 1 recorded result & sources 0.496 edit distance · raw 0.496 edit distance Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table12; thinking,280 visual tokens; separate visual-resolution configuration, not a new checkpoint. https://arxiv.org/pdf/2607.02770#page=17
RULER32K (Gemma,no thinking)	83%none effort	All 1 recorded result & sources 83% · raw 83 % Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. RULER32K. Native %. https://arxiv.org/pdf/2607.02770#page=8
RULER128K (Gemma,no thinking)	70.4%none effort	All 1 recorded result & sources 70.4% · raw 70.4 % Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. RULER128K. Native %. https://arxiv.org/pdf/2607.02770#page=8
LOFT text retrieval128K (Recall@k) (Gemma,no thinking)	50.5%none effort	All 1 recorded result & sources 50.5% · raw 50.5 % Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. LOFT text retrieval128K (Recall@k). Native %. https://arxiv.org/pdf/2607.02770#page=8
GraphWalks <128K (F1) (Gemma,no thinking)	4.1%none effort	All 1 recorded result & sources 4.1% · raw 4.1 % Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. GraphWalks <128K (F1). Native %. https://arxiv.org/pdf/2607.02770#page=8
MTOB English→Kalamang ~128K half-book (Gemma,no thinking)	15.4 chrFnone effort	All 1 recorded result & sources 15.4 chrF · raw 15.4 chrF Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. MTOB English→Kalamang ~128K half-book. Native chrF. https://arxiv.org/pdf/2607.02770#page=8
MTOB Kalamang→English ~128K half-book (Gemma,no thinking)	28.2 chrFnone effort	All 1 recorded result & sources 28.2 chrF · raw 28.2 chrF Headline · none effort · Own vendor Source/record date: 2026-10-19 Harness: Google report Table9, thinking disabled. MTOB Kalamang→English ~128K half-book. Native chrF. https://arxiv.org/pdf/2607.02770#page=8
τ²-bench (Tau-2)	24.5%unknown effort	All 1 recorded result & sources 24.5% · raw 24.5 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google official card; published average over3 domains, thinking on. Not recomputed from report domains. https://huggingface.co/google/gemma-4-E2B-it/blob/3e22461f65e89153144f8adb70e3b8c2cc9845a7/README.md
CoVoST (Gemma card aggregate)	33.47 BLEUunknown effort	All 1 recorded result & sources 33.47 BLEU · raw 33.47 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google card audio aggregate. CoVoST translation corpus BLEU.12B excludes Chinese. Distinct from report Table7/8. https://huggingface.co/google/gemma-4-E2B-it/blob/3e22461f65e89153144f8adb70e3b8c2cc9845a7/README.md
FLEURS (Gemma card aggregate error)	9.00%unknown effort	All 1 recorded result & sources 9.00% · raw 9 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google card audio aggregate. FLEURS error fraction converted to percent; WER/CER language conventions are not restated on card.12B excludes Chinese. Distinct from report Table7/8. https://huggingface.co/google/gemma-4-E2B-it/blob/3e22461f65e89153144f8adb70e3b8c2cc9845a7/README.md
CoVoST ja→English (Gemma report)	21.4 BLEUunknown effort	All 1 recorded result & sources 21.4 BLEU · raw 21.4 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; ja source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST de→English (Gemma report)	39.2 BLEUunknown effort	All 1 recorded result & sources 39.2 BLEU · raw 39.2 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; de source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST fr→English (Gemma report)	39.2 BLEUunknown effort	All 1 recorded result & sources 39.2 BLEU · raw 39.2 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; fr source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST es→English (Gemma report)	43.2 BLEUunknown effort	All 1 recorded result & sources 43.2 BLEU · raw 43.2 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; es source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST it→English (Gemma report)	40.8 BLEUunknown effort	All 1 recorded result & sources 40.8 BLEU · raw 40.8 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; it source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST ru→English (Gemma report)	46.4 BLEUunknown effort	All 1 recorded result & sources 46.4 BLEU · raw 46.4 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; ru source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST zh→English (Gemma report)	17.9 BLEUunknown effort	All 1 recorded result & sources 17.9 BLEU · raw 17.9 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; zh source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
CoVoST 7-language average (Gemma report)	35.4 BLEUunknown effort	All 1 recorded result & sources 35.4 BLEU · raw 35.4 BLEU Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; average source language; native corpus BLEU, not percent accuracy. https://arxiv.org/pdf/2607.02770#page=7
FLEURS en (WER,Gemma report)	8.00%unknown effort	All 1 recorded result & sources 8.00% · raw 8 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS ko (CER,Gemma report)	6.60%unknown effort	All 1 recorded result & sources 6.60% · raw 6.6 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription CER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS ja (CER,Gemma report)	10.70%unknown effort	All 1 recorded result & sources 10.70% · raw 10.7 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription CER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS de (WER,Gemma report)	7.60%unknown effort	All 1 recorded result & sources 7.60% · raw 7.6 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS fr (WER,Gemma report)	10.10%unknown effort	All 1 recorded result & sources 10.10% · raw 10.1 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS hi (WER,Gemma report)	10.10%unknown effort	All 1 recorded result & sources 10.10% · raw 10.1 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS es (WER,Gemma report)	4.20%unknown effort	All 1 recorded result & sources 4.20% · raw 4.2 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS it (WER,Gemma report)	4.10%unknown effort	All 1 recorded result & sources 4.10% · raw 4.1 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS pt-br (WER,Gemma report)	5.60%unknown effort	All 1 recorded result & sources 5.60% · raw 5.6 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS ru (WER,Gemma report)	8.40%unknown effort	All 1 recorded result & sources 8.40% · raw 8.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS ar (WER,Gemma report)	14.30%unknown effort	All 1 recorded result & sources 14.30% · raw 14.3 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription WER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS zh (CER,Gemma report)	18.70%unknown effort	All 1 recorded result & sources 18.70% · raw 18.7 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription CER. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
FLEURS average (mixed WER/CER macro error,Gemma report)	9.00%unknown effort	All 1 recorded result & sources 9.00% · raw 9 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: Google report audio Table7; transcription mixed WER/CER macro error. Source fraction retained. https://arxiv.org/pdf/2607.02770#page=7
MMMLU (Gemma launch figure label)	60%unknown effort	All 1 recorded result & sources 60% · raw 60 % Headline · unknown effort · Own vendor Source/record date: 2026-10-19 Harness: literal Google launch figure label; benchmark-name conflict with full card/report; this cell is not used as ordinary MMLU-Pro or MMMLU. https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/

Independent evaluators

Evaluator harnesses are distinct from vendor measurements. Missing coverage is not a failed test.

Independent evaluator headline scores; expand evidence for every record
Benchmark / evaluator	Headline score	Evidence
GDPval-AA v2.1 Elo · Artificial Analysis	-188unknown effort	All 1 recorded result & sources -188 · raw -188 Elo Headline · unknown effort · Independent evaluator Source/record date: 2026-10-20 Manually observed public GDPval-AA v2.1 leaderboard table, full 298-configuration pool checked 20 October 2026. Exact source configuration: Gemma 4 E2B (Reasoning); creator: Google; source rank 284; source release-month label Apr 2026 is model release metadata, not execution date. Published rounded native Elo −188; literal displayed CI -25 / +25. Native Elo is not a percentage and is not the normalized (Elo−500)/2000 chart quantity. This versioned metric is separate from older generic GDPval-AA Elo and vendor GDPval/GDPval-AA results; no old evidence is overwritten. Execution date, inference precision, unprinted effort budgets and task costs are unpublished in this captured table. Elo is relative to the dated comparison pool; drift from an earlier observation does not prove a new evaluation run. Source says Reasoning (on), not an API effort level; reasoning effort remains unknown. Headline: highest published native Elo for this exact approved model; tied values prefer existing featured effort then source order. https://artificialanalysis.ai/evaluations/gdpval-aa
Intelligence Index (estimated) · Artificial Analysis	7.8unknown effort	All 2 recorded results & sources 7.8 · raw 7.77445525187484 index Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 6.5 · raw 6.47543520107148 index Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
AA-Omniscience Index · Artificial Analysis	-23.6unknown effort	All 2 recorded results & sources -23.6 · raw -23.583333333333332 index Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b -60.2 · raw -60.18333333333333 index Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
AA-Omniscience accuracy · Artificial Analysis	6.6%unknown effort	All 2 recorded results & sources 6.6% · raw 6.633333333333333 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 8.2% · raw 8.15 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
AA-Omniscience hallucination rate · Artificial Analysis	32.4%unknown effort	All 2 recorded results & sources 32.4% · raw 32.36344162799 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 74.4% · raw 74.39666122300854 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
GDPval-AA Elo · Artificial Analysis	-187.7unknown effort	All 1 recorded result & sources -187.7 · raw -187.7 Elo Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b
Terminal-Bench 2.1 (AA) · Artificial Analysis	0.4%unknown effort	All 1 recorded result & sources 0.4% · raw 0.374531835205993 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b
Terminal Bench Hard (AA) · Artificial Analysis	3%unknown effort	All 2 recorded results & sources 3% · raw 3.03030303030303 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 2.3% · raw 2.27272727272727 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
AA-LCR · Artificial Analysis	16.3%unknown effort	All 2 recorded results & sources 16.3% · raw 16.3333333333333 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 16.3% · raw 16.3333333333333 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
IFBench (AA) · Artificial Analysis	38%unknown effort	All 2 recorded results & sources 38% · raw 38.0272108843537 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 33.6% · raw 33.605442176870696 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
Humanity's Last Exam (AA harness) · Artificial Analysis	4.8%unknown effort	All 2 recorded results & sources 4.8% · raw 4.77293790546803 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 4.7% · raw 4.72659870250232 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
GPQA Diamond (AA) · Artificial Analysis	43.3%unknown effort	All 2 recorded results & sources 43.3% · raw 43.3333333333333 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 40.5% · raw 40.5050505050505 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
CritPt (AA) · Artificial Analysis	0%unknown effort	All 2 recorded results & sources 0% · raw 0 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 0% · raw 0 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
MMMU-Pro (AA) · Artificial Analysis	44.6%unknown effort	All 2 recorded results & sources 44.6% · raw 44.5664739884393 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 41.8% · raw 41.7919075144509 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning
τ²-Telecom (AA) · Artificial Analysis	20.8%unknown effort	All 2 recorded results & sources 20.8% · raw 20.7602339181287 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Reasoning enabled without a published common API effort label. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b 22.2% · raw 22.2222222222222 % Alternative · none effort · Independent evaluator Source/record date: 2026-10-19 Independent AA exact model page explicitly links https://huggingface.co/google/gemma-4-E2B-it. Explicit non-reasoning configuration. Index v4.3.2; inference precision, execution date and exact budget unpublished. Linking reference weights does not prove BF16 inference. Costs with unavailable/zero model prices are not imported as free tariffs. https://artificialanalysis.ai/models/gemma-4-e2b-non-reasoning

Read how we select and source scores or the comparison guide.