Alibaba / QwenQwen
Qwen 3 Coder 480B
Alibaba / Qwen · Qwen. Released Jul 22, 2025.
Qwen 3 Coder 480B is a model from Alibaba / Qwen in the Qwen family, released Jul 22, 2025. evals.report tracks 11 reported Qwen 3 Coder 480B benchmark scores across SWE-bench Pro, WeirdML, Aider Polyglot, MMLU-Pro, GDPval, LiveCodeBench, SciCode, WebDev Arena, and 3 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
Open11 results
Benchmark results 11
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| SWE-bench Pro | Coding | 38.70% | % resolved | Official | Jul 22, 2025 | Details |
| WeirdML | Coding | 41.2% | average accuracy | Official | Jul 22, 2025 | Details |
| Aider Polyglot | Coding | 61.8% | % correct | Unverified | Jul 22, 2025 | Details |
| MMLU-Pro | Reasoning | 78.8% | accuracy | Verified | Jul 22, 2025 | Details |
| GDPval | Agents | 506 | Elo | Official | Jul 22, 2025 | Details |
| LiveCodeBench | Coding | 58.5% | Pass@1 | Unverified | Jul 22, 2025 | Details |
| SciCode | Coding | 35.9% | accuracy | Unverified | Jul 22, 2025 | Details |
| WebDev Arena | Chat preference | 1282 | Elo | Verified | Jul 22, 2025 | Details |
| Design Arena | Chat preference | 1197 | Elo | Verified | Jul 22, 2025 | Details |
| MCP-Universe | Tool use | 22.94% | Overall Success Rate | Verified | Jul 22, 2025 | Details |
| GSO: Software Optimization Benchmark for SWE-Agents | Coding | 4.9% | Opt@1 | Official | Jul 22, 2025 | Details |