Alibaba
Qwen2.5 Turbo is an Alibaba Qwen model profile, from a broad open and commercial model family covering general chat, coding, reasoning, multimodal, and compact deployment use cases. The local benchmark view helps place this exact Qwen variant among nearby open-weight and hosted competitors.
Qwen model releasesRank #400 across 526
No rank
Rank #21 across 357
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 12.0 | #400 |
| MMLU-Pro | reasoning | 63.3% | #260 |
| GPQA | reasoning | 41.0% | #410 |
| Humanity's Last Exam | reasoning | 4.2% | #418 |
| LiveCodeBench |
| coding |
| 16.3% |
| #285 |
| SciCode | coding, reasoning | 15.3% | #418 |
| MATH-500 | math | 80.5% | #108 |
| AIME | math | 12.0% | #121 |
| Output Speed | speed | 66.4 tok/s | #192 |
| Time to First Token | speed | 1.19s | #156 |
| Blended Price | cost | $0.088/M | #21 |
| Input Price | cost | $0.050/M | #25 |
| Output Price | cost | $0.200/M | #31 |
| Value Index | cost, overall | 136.4 | #36 |