Which is better, Gemma 4 E2B or Qwen3.5 4B?

Gemma 4 E2B vs Qwen3.5 4B

47 SHARED BENCHMARKS

Across 47 shared benchmarks, Gemma 4 E2B scores higher on 4 and Qwen3.5 4B on 42, with 1 level. The widest gap is Terminal-Bench Hard, where Qwen3.5 4B scores 18.2 against 3.

GOOGLEVSALIBABA47 SHARED4–42 HEAD-TO-HEAD

BenchmarkGemma 4 E2BMARGINQwen3.5 4B

AA Agentic Index7.436.3

AA Intelligence1020

AA-LCR16.363

AA-Omniscience-23.6-58.4

aime_202637.599

AIME25 no tools26.354.3

Artificial Analysis Coding Index8.322.6

BFCL v356.471.1

bfcl_v43754

BrowseComp+ (OpenClaw)8.324.5

Claw-Eval Avg53.162.3

critpt00

gdpval08.3

GPQA Diamond43.477.1

HLE4.89.9

IFBench3859.2

ifeval8389.8

LiveCodeBench v64455.8

LiveCodeBenchV6 no tools54.960.9

MATH500 (Pass@1)6480.8

MathVista62.163.6

MM-IFEval65.663.1

MMBench64.287.1

MME54.979.5

MMLU-Pro6079.1

mmmlu67.476.1

MMMU (val) (Pass@1)41.150.3

MMMU-Pro44.665.4

MMStar57.975.3

Multilingual MMBench62.877

Multilingual MMMB73.382

OCRBench71.986.9

OmniScience Accuracy8.215.1

OmniScience Non-Hallucination67.613.4

PinchBench44.271.3

RealWorldQA6076.3

scicode20.918.3

SEED-Bench (image)71.476.1

SimpleVQA27.347.8

t2-bench29.479.9

Tau² Telecom22.487.7

TauBench V3 - Banking4.56.8

Terminal-Bench 2.10.425.8

Terminal-Bench Hard318.2

τ²-Bench (Retail)18.971.9

τ²-Bench Telecom (AA run)22.292.1

τ³-Bench Banking3.45.5

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.

Compare Gemma 4 E2B withALL PAIRINGS →

GoogleGemini 3.1 Pro Gemini 3 Pro Gemma 4 26B A4B Gemini 2.5 Pro Gemini 1.5 Pro Gemma 4 31B

Across the fieldGemini 3.6 Flash Nemotron 3 Super 120B A12B GPT-5.5 Claude Opus 4.8 GPT-4o GPT-5.6 Sol GPT-5.4 GPT-5.2

Compare Qwen3.5 4B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen2.5 Instruct 72B Qwen3.5 27B Qwen3.6 Plus

Across the fieldGPT-4o mini Qwen3.6 35B A3B Gemini 3.5 Flash GPT-6 Astra MiniMax M2.7 Qwen3.5 2B Qwen3.6 27B Qwen3.7 Max