Which is better, Claude Sonnet 4 or Qwen3.5 2B?

Claude Sonnet 4 vs Qwen3.5 2B

28 SHARED BENCHMARKS

Across 28 shared benchmarks, Claude Sonnet 4 scores higher on 27 and Qwen3.5 2B on 1. The widest gap is Terminal-Bench Hard, where Claude Sonnet 4 scores 31.1 against 3.8.

ANTHROPICVSALIBABA28 SHARED27–1 HEAD-TO-HEAD

BenchmarkClaude Sonnet 4MARGINQwen3.5 2B

AA Agentic Index39.20.8

AA Intelligence256.9

AA-LCR70.325.6

AA-Omniscience0.2-60.2

AIME 202443.417

AIME 20257419

Artificial Analysis Coding Index37.62.9

critpt1.10

gdpval31.20

GPQA Diamond7851.6

HLE10.75

IFBench5541.3

ifeval87.678.6

math_500_em9481.5

mmlu_redux93.679.6

MMLU-Pro8466.5

mmmlu86.563.1

MMMU-Pro62.443

multichallenge57.133.7

OmniScience Accuracy22.78.1

OmniScience Non-Hallucination70.925.8

scicode407.2

supergpqa55.737.5

TauBench V3 - Banking16.72.5

Terminal-Bench 2.136.33

Terminal-Bench Hard31.13.8

τ²-Bench Telecom (AA run)64.681.6

τ³-Bench13.86.2

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 Claude Sonnet 4 withALL PAIRINGS →

AnthropicClaude Opus 4.8 Claude Opus 4.6 Claude Sonnet 4.5 Claude 3.5 Sonnet Claude Fable 5 Claude Opus 4.5

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

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 fieldClaude Sonnet 5 Muse Spark 1.1 GLM-5.1 GLM-5.2 o4-mini Claude Sonnet 4.6 o1 Gemma 4 31B