Which is better, Qwen3.5 2B or Qwen3.6 35B A3B?

Qwen3.5 2B vs Qwen3.6 35B A3B

27 SHARED BENCHMARKS

Across 27 shared benchmarks, Qwen3.5 2B scores higher on 0 and Qwen3.6 35B A3B on 27. The widest gap is Terminal-Bench Hard, where Qwen3.6 35B A3B scores 34.8 against 3.8.

ALIBABAVSALIBABA27 SHARED0–27 HEAD-TO-HEAD

BenchmarkQwen3.5 2BMARGINQwen3.6 35B A3B

AA Agentic Index0.819.2

AA Intelligence6.918.8

AA-LCR25.671.7

AA-Omniscience-60.2-22.2

Artificial Analysis Coding Index2.941.9

C-Eval73.290

critpt00.3

gdpval024.5

GPQA Diamond51.686

HallusionBench65.569.8

HLE522.2

IFBench41.364.4

MathVista48.786.4

mmlu_redux79.693.3

MMLU-Pro66.585.6

MMMU-Pro4375.3

OmniScience Accuracy8.118.8

OmniScience Non-Hallucination25.849.5

RealWorldQA65.185.3

scicode7.236.6

SimpleVQA35.258.9

supergpqa37.564.7

TauBench V3 - Banking2.59.3

Terminal-Bench 2.1344.9

Terminal-Bench Hard3.834.8

τ²-Bench Telecom (AA run)81.695.3

τ³-Bench6.267.2

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 Qwen3.5 2B 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 fieldGemini 3.5 Flash GPT-6 Astra MiniMax M2.7 Qwen3.6 27B Qwen3.7 Max Qwen3.8 Max Preview Qwen3.7 Plus Preview Claude Opus 4

Compare Qwen3.6 35B A3B 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 fieldQwen3.6 27B Qwen3.7 Max Qwen3.8 Max Preview Qwen3.7 Plus Preview Claude Opus 4 Qwen3.5 4B Gemini 3.6 Flash Nemotron 3 Super 120B A12B