Which is better, Gemini 3 Pro or Qwen3.5 4B?

Gemini 3 Pro vs Qwen3.5 4B

29 SHARED BENCHMARKS

Across 29 shared benchmarks, Gemini 3 Pro scores higher on 27 and Qwen3.5 4B on 2. The widest gap is HLE, where Gemini 3 Pro scores 45.8 against 9.9. Qwen3.5 4B is the cheaper of the two on tracked API pricing ($0.03 against $2.00 per million input tokens).

GOOGLEVSALIBABA29 SHARED27–2 HEAD-TO-HEAD

BenchmarkGemini 3 ProMARGINQwen3.5 4B

AA Agentic Index5236.3

AA Intelligence2820

AA-LCR7663

AA-Omniscience15.3-58.4

aime_202691.799

AIME25 no tools9654.3

Artificial Analysis Coding Index46.522.6

critpt9.10

gdpval34.28.3

GPQA Diamond91.977.1

HLE45.89.9

hmmt_feb_202686.484.1

IFBench70.459.2

LiveCodeBench v690.755.8

longbench_v268.250

MathVista89.863.6

MMLU-Pro90.179.1

mmmlu91.876.1

MMMU-Pro8165.4

multichallenge65.749

OCRBench90.386.9

OmniScience Accuracy55.815.1

OmniScience Non-Hallucination1013.4

scicode56.118.3

SimpleVQA69.747.8

t2-bench85.479.9

Terminal-Bench Hard56.918.2

τ²-Bench (Retail)85.371.9

τ²-Bench Telecom (AA run)9892.1

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. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.

Compare Gemini 3 Pro withALL PAIRINGS →

GoogleGemini 3.1 Pro Gemma 4 26B A4B Gemini 2.5 Pro Gemini 1.5 Pro Gemma 4 31B Gemma 4 E2B

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 fieldClaude Fable 5 GPT-5 Claude Opus 4.5 DeepSeek-V4-Pro DeepSeek-V3 Kimi K2.6 Kimi K2.5 o3