Which is better, Qwen2.5 Instruct 72B or Qwen3.5 2B?

Qwen2.5 Instruct 72B vs Qwen3.5 2B

21 SHARED BENCHMARKS

Across 21 shared benchmarks, Qwen2.5 Instruct 72B scores higher on 13 and Qwen3.5 2B on 7, with 1 level. The widest gap is Artificial Analysis Coding Index, where Qwen2.5 Instruct 72B scores 11.9 against 2.9.

ALIBABAVSALIBABA21 SHARED13–7 HEAD-TO-HEAD

BenchmarkQwen2.5 Instruct 72BMARGINQwen3.5 2B

AA Intelligence106.9

AA-LCR20.325.6

AA-Omniscience-52.2-60.2

AIME 202423.317

Artificial Analysis Coding Index11.92.9

C-Eval89.273.2

critpt00

GPQA Diamond49.151.6

HLE4.25

IFBench36.941.3

ifeval87.278.6

longbench_v239.438.7

math_500_em8081.5

mmlu_redux86.879.6

MMLU-Pro71.666.5

mmmlu74.863.1

OmniScience Accuracy17.68.1

OmniScience Non-Hallucination15.325.8

scicode26.77.2

Terminal-Bench Hard4.53.8

τ²-Bench Telecom (AA run)34.581.6

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 Qwen2.5 Instruct 72B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen3.5 27B Qwen3.6 Plus Qwen3.6 35B A3B

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 Qwen3.5 27B Qwen3.6 Plus Qwen3.6 35B A3B

Across the fieldGemini 1.5 Pro Claude Sonnet 4 Claude Sonnet 5 Muse Spark 1.1 GLM-5.1 GLM-5.2 o4-mini Claude Sonnet 4.6