Alibaba
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the...
Qwen model releasesRank #303 across 566
No rank
Rank #176 across 371
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 | 13.6 | #303 |
| Artificial Analysis Math Index | math | 29.0 | #188 |
| MMLU-Pro | reasoning | 70.6% | #225 |
| GPQA | reasoning | 51.6% | #387 |
| Humanity's Last Exam |
| reasoning |
| 4.0% |
| #476 |
| LiveCodeBench | coding | 40.3% | #179 |
| SciCode | coding, reasoning | 27.8% | #339 |
| MATH-500 | math | 89.3% | #75 |
| AIME | math | 29.7% | #86 |
| Output Speed | speed | 97.6 tok/s | #145 |
| Time to First Token | speed | 1.38s | #196 |
| Blended Price | cost | $0.900/M | #176 |
| Input Price | cost | $0.450/M | #164 |
| Output Price | cost | $2.25/M | #171 |
| Value Index | cost, overall | 15.1 | #198 |