Alibaba
The Qwen3 series VL models has been comprehensively upgraded in areas such as visual coding and spatial perception. Its visual perception and recognition capabilities have significantly improved, supporting the understanding of ultra-long videos, and its OCR functionality has undergone a major enhancement.
Qwen model releasesRank #235 across 526
Rank #247 across 436
Rank #166 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 | 20.8 | #235 |
| Artificial Analysis Coding Index | coding | 16.5 | #247 |
| Artificial Analysis Math Index | math | 70.7 | #95 |
| MMLU-Pro | reasoning | 82.3% | #71 |
| reasoning |
| 71.2% |
| #211 |
| Humanity's Last Exam | reasoning | 6.3% | #251 |
| LiveCodeBench | coding | 59.4% | #117 |
| SciCode | coding, reasoning | 35.9% | #195 |
| Output Speed | speed | 48.1 tok/s | #246 |
| Time to First Token | speed | 1.17s | #151 |
| Blended Price | cost | $0.700/M | #166 |
| Input Price | cost | $0.300/M | #164 |
| Output Price | cost | $1.90/M | #176 |
| Value Index | cost, overall | 29.7 | #183 |