Qwen3-14B ranks first on MathIF.
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 | |
|---|---|---|---|---|
| MathIF | 50.7% | #1 of 12 | Official leaderboard | |
| ALRAGE | 79.7% | #6 of 39 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 5.40% | #17 of 104 | Official leaderboard | |
| Arena-Hard-Auto | 91.7% | #8 of 39 | Model card | |
| RULER | 94.6% | #9 of 38 | Model card | |
| MadinahQA | 52.8% | #13 of 39 | Official leaderboard | |
| AlignBench | 8.56 | #5 of 15 | Model card | |
| Multi-IF | 74.8% | #10 of 28 | Model card | |
| MLogiQA | 71.1% | #3 of 8 | Model card | |
| ArbMMLU-HT | 47.8% | #16 of 39 | Official leaderboard | |
| Open Arabic LLM Leaderboard v2 | 53.8% | #19 of 39 | Official leaderboard | |
| AlGhafa | 57.1% | #20 of 39 | Official leaderboard | |
| Fiction.LiveBench | 62.5% | #30 of 57 | Epoch AI | |
| IFEval | 85.4% | #71 of 126 | Model card | |
| ArabicMMLU | 44.6% | #22 of 39 | Official leaderboard | |
| MT-AIME2024 | 73.3% | #4 of 7 | Model card | |
| Creative Writing v3 | 80.3 | #7 of 12 | Model card | |
| WritingBench | 7.80 | #9 of 14 | Model card | |
| SEA-HELM | 64.8 | #40 of 62 | Official leaderboard | |
| MMLU-Redux | 88.6% | #25 of 38 | Model card | |
| INCLUDE | 71.7% | #17 of 25 | Model card | |
| C-Eval | 86.2% | #26 of 38 | Model card | |
| MMMLU | 77.9% | #46 of 62 | Model card | |
| Vectara Hallucination Leaderboard (HHEM) | 2.20% | #123 of 148 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 35.9 | #103 of 119 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AutoLogi | 89.2% | #2 of 8 | Model card | |
| MATH-500 | 96.8% | #23 of 67 | Model card | |
| ZebraLogic | 88.5% | #10 of 23 | Model card | |
| AIME 2024 | 79.3% | #43 of 87 | Model card | |
| OTIS Mock AIME 2024-2025 | 66.4% | #108 of 202 | Epoch AI | |
| GPQA Diamond | 63.8% | #123 of 219 | Epoch AI | |
| DTBench | 40% | #126 of 178 | Official leaderboard | |
| PolyMath | 45.8% | #14 of 18 | Model card | |
| AIME 2025 | 70.4% | #112 of 143 | Model card | |
| CritPt | 0% | #135 of 160 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 63.5% | #20 of 44 | Model card | |
| AutoCodeBench | 37.6% | #24 of 34 | Official leaderboard | |
| Codeforces rating | 1,766 | #30 of 41 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BFCL V4 | 41% | #28 of 65 | Official leaderboard | |
| Berkeley Function Calling Leaderboard (BFCL) | 70.4% | #21 of 47 | Model card | |
| SnakeBench | 20 | #132 of 183 | Official leaderboard | |
| LisanBench | 58.3 | #120 of 133 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AraTrust | 57.5% | #28 of 39 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 71.3% | #11 of 27 | Model card | |
| ACCoRD | 48.4 | #114 of 150 | Official leaderboard | |
| LMCA | 18.2% | #108 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 28.6 | #116 of 149 | 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. 25 of them aren't in the launch post.