Qwen2.5-Coder 32B Instruct
llmCode-specialized member of the Qwen2.5 family that Alibaba positions as matching GPT-4o's coding ability among open models.
Metadata
| org | Alibaba |
| released | 2024-11-12 |
| params | 32B |
| license | apache-2.0 |
| kind | llm |
| hf id | Qwen/Qwen2.5-Coder-32B-Instruct |
Benchmarks
| MMLU | 79.1 | source ↗ |
| MMLU-Pro | 62.3 | source ↗ |
| GPQA | 41.8 | source ↗ |
| HumanEval | 92.7 | source ↗ |
Lineage
No recorded lineage — root or standalone model.
view in full graph →