Qwen 2.5 Coder 32B

#26Open Source

Alibaba/Qwen

State-of-the-art open-source coding model. Strong HumanEval score.

Official page →

📖Context128K tokens

🚀Speed90 tok/s

💵Input price-

💸Output price-

🧠Parameters32B

Benchmarks

Gray bar = dataset average

MMLU57 academic subjects

80.0%avg 81.8

GPQA DiamondPhD-level science questions

42.0%avg 50.4

HumanEvalPython code generation

92.3%avg 79.9

SWE-Bench VerifiedReal GitHub engineering tasks

30.0%avg 32.4

Composite ScoreScore = MMLU×20% + GPQA×30% + HumanEval×25% + SWE-Bench×25%59.2

HF Open LLM Leaderboard

IFEval · BBH · MATH · GPQA · MuSR · MMLU-Pro

Average

39.9

View on HF →

IFEvalInstruction following

72.7%avg 79.9

BBHBig Bench Hard

52.3%avg 48.7

MATH Lvl 5Competition mathematics

49.5%avg 37.8

GPQAGraduate-level science Q&A

13.2%avg 12.9

MuSRMultistep soft reasoning

13.7%avg 10.7

MMLU-ProProfessional knowledge

37.9%avg 40.0

More from Alibaba/Qwen

Qwen3 235B A22B Thinking 2507

$0.100/1M→

API Providers

Details

Provider: Alibaba/Qwen
Parameters: 32B
Context: 128K tokens
Speed: 90 tok/s
Open Source: Yes
License: Open (varies)
Released: -

Reviews

No reviews yet

← View all 376 models