Qwen2.5 Coder 32B Instruct
About
Instruction-optimized 32B code flagship for production systems requiring top-tier code reasoning, generation, and multi-file analysis.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution
Providers(5)
Compare all →| Provider | Input (per 1M) | Output (per 1M) | Type | |
|---|---|---|---|---|
| OpenRouter | $0.66 | $1 | Serverless | |
| Fireworks AI | $0.9 | $0.9 | Serverless | |
| NVIDIA NIM | — | — | Serverless | |
| Arcee AI | $0.4 | $1.2 | Serverless | |
| SiliconFlow | $0.18 | $0.18 | Serverless |
Benchmark Scores(3)
| Benchmark | Score | Version | Source |
|---|---|---|---|
| HumanEval | 92.7 | 2024-11 | Qwen Team |
| Aider Polyglot | 16.4 | 2026-04 | https://aider.chat/docs/leaderboards |
| BigCodeBench | 49.0 | 2025-01 (Instruct Pass@1) | https://bigcode-bench.github.io/results.json |
Specifications
FamilyQwen2.5 Coder
Released2024-11-12
Parameters32B
ArchitectureDecoder Only
Specializationcoding
Trainingfinetuning
Fine-tuninginstruct
