VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which is better, Gemma 4 E2B or Qwen3.5 122B A10B?
Across 30 shared benchmarks, Gemma 4 E2B scores higher on 2 and Qwen3.5 122B A10B on 28. The widest gap is HLE, where Qwen3.5 122B A10B scores 47.5 against 4.8.

Gemma 4 E2B vs Qwen3.5 122B A10B

Across 30 shared benchmarks, Gemma 4 E2B scores higher on 2 and Qwen3.5 122B A10B on 28. The widest gap is HLE, where Qwen3.5 122B A10B scores 47.5 against 4.8.

GooglevsAlibaba30 shared benchmarks228 head-to-head
BenchmarkGemma 4 E2BQwen3.5 122B A10B
AA Agentic Index7.421.3
AA Intelligence1032.8
AA-LCR1770.3
AA-Omniscience-23.6-41.5
Artificial Analysis Coding Index8.345.7
bfcl_v43772.2
critpt00.9
gdpval024.3
GPQA Diamond43.486.6
HallusionBench0.667.6
HLE4.847.5
IFBench3876.1
ifeval82.993.4
LiveCodeBench v64478.9
mathvision52.486.2
MathVista62.187.4
MMLU-Pro6086.7
mmmlu67.486.7
MMMU-Pro44.676.9
MMStar57.982.9
OCRBench0.792.1
OmniScience Accuracy8.224.4
OmniScience Non-Hallucination67.612.9
RealWorldQA56.285.1
scicode20.942
t2-bench29.479.5
TauBench V3 - Banking4.515.3
Terminal-Bench 2.10.447.6
Terminal-Bench Hard331.1
τ²-Bench Telecom (AA run)22.293.6

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.