Qwen3-30B-A3B places in the top three on MT-AIME2024.
Qwen Released 28 Apr 2025 Qwen Apr 2025 Open weights
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LLM Confabulation Leaderboard | 12.3% | #7 of 58 | Official leaderboard | |
| Arena-Hard-Auto | 91% | #9 of 39 | Model card | |
| AlignBench | 8.70 | #4 of 15 | Model card | |
| RULER | 91.6% | #15 of 38 | Model card | |
| MT-AIME2024 | 73.9% | #3 of 7 | Model card | |
| IFEval | 86.5% | #60 of 126 | Model card | |
| Multi-IF | 72.2% | #14 of 28 | Model card | |
| MLogiQA | 70.1% | #4 of 8 | Model card | |
| MMLU-Redux | 89.5% | #22 of 38 | Model card | |
| LMArena (Elo) | 1,327 | #199 of 333 | Official leaderboard | |
| C-Eval | 86.6% | #24 of 38 | Model card | |
| Lech Mazur Creative Writing | 7.53 | #28 of 44 | Epoch AI | |
| INCLUDE | 71.9% | #16 of 25 | Model card | |
| Arena-Hard v2.0 | 33.9% | #15 of 23 | Official leaderboard | |
| LongBench v2 | 42.5% | #22 of 33 | Official leaderboard | |
| MMMLU | 78.4% | #43 of 62 | Model card | |
| WritingBench | 7.70 | #10 of 14 | Model card | |
| Creative Writing v3 | 79.1 | #9 of 12 | Model card | |
| EQ-Bench 3 | 917 | #66 of 78 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 35.6 | #105 of 119 | Official leaderboard | |
| Fiction.LiveBench | 40.6% | #52 of 57 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH-500 | 98% | #8 of 67 | Model card | |
| ZebraLogic | 89.5% | #6 of 23 | Model card | |
| AIME 2024 | 80.4% | #41 of 87 | Model card | |
| Misguided Attention | 43.6% | #24 of 45 | Official leaderboard | |
| OTIS Mock AIME 2024-2025 | 62.8% | #112 of 202 | Epoch AI | |
| GPQA Diamond | 61.7% | #125 of 219 | Epoch AI | |
| HiPhO | 47.1% | #21 of 35 | Official leaderboard | |
| AutoLogi | 88.7% | #5 of 8 | Model card | |
| HMMT 2025 | 50.8% | #38 of 53 | Official leaderboard | |
| PolyMath | 46.1% | #13 of 18 | Model card | |
| AIME 2025 | 70.9% | #111 of 143 | Model card | |
| DTBench | 33.8% | #141 of 178 | Official leaderboard | |
| SMT 2025 | 67.9% | #35 of 40 | Official leaderboard | |
| BRUMO 2025 | 77.5% | #38 of 42 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 62.6% | #22 of 44 | Model card | |
| Codeforces rating | 1,974 | #27 of 41 | Model card | |
| WeirdML | 29.8% | #99 of 112 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Berkeley Function Calling Leaderboard (BFCL) | 69.1% | #25 of 47 | Model card | |
| Step Race | 2 | #32 of 59 | Official leaderboard | |
| SnakeBench | 16 | #162 of 183 | Official leaderboard | |
| Elimination Game | 2 | #55 of 56 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 74.3% | #9 of 27 | Model card | |
| ACCoRD | 45.0 | #121 of 150 | Official leaderboard | |
| Conceptual Reasoning Index | 25.2 | #124 of 149 | Official leaderboard | |
| LMCA | 15.8% | #117 of 139 | Epoch AI | |
| Political Science LLM Benchmark | 59.3% | #29 of 29 | Official leaderboard |
Qwen's models in color, against every other model on the Epoch Capabilities Index, by release date.
Every result in the longer documents the lab published with it. 24 of them aren't in the launch post.