Alibaba / QwenQwen
Qwen3 Max
Alibaba / Qwen · Qwen. Released Sep 5, 2025.
Qwen3 Max is a model from Alibaba / Qwen in the Qwen family, released Sep 5, 2025. evals.report tracks 15 reported Qwen3 Max benchmark scores across SimpleQA Verified, GPQA Diamond, AIME (OTIS Mock), LiveCodeBench Pro, PostTrainBench, Artificial Analysis Intelligence Index, Epoch Capabilities Index, MMLU-Pro, and 7 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
15 results
Benchmark results 15
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| SimpleQA Verified | Other | 67.5% | accuracy | Official | Sep 5, 2025 | Details |
| GPQA Diamond | Reasoning | 72.6% | accuracy | Official | Sep 5, 2025 | Details |
| AIME (OTIS Mock) | Reasoning | 73.3% | accuracy | Official | Sep 5, 2025 | Details |
| LiveCodeBench Pro | Coding | 1226 | Codeforces Elo | Official | Sep 5, 2025 | Details |
| PostTrainBench | Agents | 7.42% | weighted average score | Official | Sep 5, 2025 | Details |
| Artificial Analysis Intelligence Index | Reasoning | 31.4 | Index | Unverified | Sep 5, 2025 | Details |
| Epoch Capabilities Index | Reasoning | 144.9 | Index | Official | Sep 5, 2025 | Details |
| MMLU-Pro | Reasoning | 84.1% | accuracy | Verified | Sep 5, 2025 | Details |
| GDPval | Agents | 1038 | Elo | Official | Sep 5, 2025 | Details |
| LiveCodeBench | Coding | 76.7% | Pass@1 | Unverified | Sep 5, 2025 | Details |
| SciCode | Coding | 43.1% | accuracy | Unverified | Sep 5, 2025 | Details |
| Global-MMLU | Reasoning | 83.3% | accuracy | Unverified | Sep 5, 2025 | Details |
| Design Arena | Chat preference | 1165 | Elo | Verified | Sep 5, 2025 | Details |
| SuperGPQA | Reasoning | 65.1% | accuracy | Unverified | Sep 5, 2025 | Details |
| Vibe Code Bench | Coding | 3.51% | Overall accuracy | Verified | Sep 5, 2025 | Details |