Which is better, GPT-4o or Qwen3.5 2B?

GPT-4o vs Qwen3.5 2B

33 SHARED BENCHMARKS

Across 33 shared benchmarks, GPT-4o scores higher on 25 and Qwen3.5 2B on 6, with 2 level. The widest gap is Artificial Analysis Coding Index, where GPT-4o scores 24.2 against 2.9.

OPENAIVSALIBABA33 SHARED25–6 HEAD-TO-HEAD

BenchmarkGPT-4oMARGINQwen3.5 2B

AA Agentic Index9.70.8

AA Intelligence11.26.9

AA-LCR5325.6

AA-Omniscience-10.5-60.2

AIME 20249.317

AIME 202511.719

Artificial Analysis Coding Index24.22.9

C-Eval7673.2

critpt00

gdpval00

GPQA Diamond70.151.6

HLE5.35

IFBench3641.3

ifeval84.378.6

longbench_v248.138.7

math_500_em74.681.5

MathVista63.848.7

mmlu_prox61.152.3

mmlu_redux8879.6

MMLU-Pro74.766.5

mmmlu81.463.1

MMMU (val) (Pass@1)69.144.1

MMMU-Pro59.943

MMStar63.955.1

multichallenge40.333.7

OCRBench80.686.1

OmniScience Accuracy23.78.1

OmniScience Non-Hallucination62.125.8

RealWorldQA75.465.1

scicode33.47.2

supergpqa42.437.5

Terminal-Bench Hard8.33.8

τ²-Bench Telecom (AA run)28.981.6

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.

Compare GPT-4o withALL PAIRINGS →

OpenAIGPT-5.5 GPT-5.6 Sol GPT-5.4 GPT-5.2 GPT-5 o3

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 fieldGPT-5.6 Sol Gemini 3.1 Pro GPT-5.4 GPT-5.2 Claude Opus 4.6 Claude Sonnet 4.5 Claude 3.5 Sonnet Gemini 3 Pro