Qwen3-4B places in the top three on MathIF.
Qwen Released 29 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 | 44% | #2 of 12 | Official leaderboard | |
| MadinahQA | 54% | #8 of 39 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 5.70% | #22 of 104 | Official leaderboard | |
| Open Arabic LLM Leaderboard v2 | 62.9% | #10 of 39 | Official leaderboard | |
| ArbMMLU-HT | 53.7% | #11 of 39 | Official leaderboard | |
| ArabicMMLU | 59.8% | #12 of 39 | Official leaderboard | |
| AlGhafa | 68.5% | #13 of 39 | Official leaderboard | |
| ALRAGE | 76.3% | #15 of 39 | Official leaderboard | |
| AlignBench | 8.30 | #7 of 15 | Model card | |
| Arena-Hard-Auto | 76.6% | #20 of 39 | Model card | |
| Multi-IF | 66.3% | #19 of 28 | Model card | |
| Slop Score | 62.2 | #12 of 17 | Official leaderboard | |
| IFEval | 81.9% | #92 of 126 | Model card | |
| RULER | 85.2% | #28 of 38 | Model card | |
| MLogiQA | 65.9% | #6 of 8 | Model card | |
| Vectara Hallucination Leaderboard (HHEM) | 2.70% | #114 of 148 | Official leaderboard | |
| C-Eval | 77.5% | #31 of 38 | Model card | |
| MMMLU | 69.8% | #52 of 62 | Model card | |
| INCLUDE | 61.8% | #21 of 25 | Model card | |
| MMLU-Redux | 83.7% | #32 of 38 | Model card | |
| MT-AIME2024 | 60.7% | #6 of 7 | Model card | |
| Arena-Hard v2.0 | 15% | #20 of 23 | Official leaderboard | |
| Creative Writing v3 | 61.1 | #11 of 12 | Model card | |
| WritingBench | 7.35 | #13 of 14 | Model card | |
| Longform Creative Writing (EQ-Bench) | 17.0 | #119 of 119 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH-500 | 97% | #20 of 67 | Model card | |
| AIME 2024 | 73.8% | #53 of 87 | Model card | |
| GPQA Diamond | 52.3% | #141 of 219 | Epoch AI | |
| ZebraLogic | 81% | #15 of 23 | Model card | |
| AutoLogi | 87.9% | #6 of 8 | Model card | |
| AIME 2025 | 65.6% | #118 of 143 | Model card | |
| PolyMath | 40% | #16 of 18 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 54.2% | #34 of 44 | Model card | |
| StructEval | 67% | #23 of 28 | Official leaderboard | |
| Codeforces rating | 1,671 | #35 of 41 | Model card | |
| AutoCodeBench | 24.3% | #30 of 34 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LisanBench | 263.3 | #83 of 133 | Official leaderboard | |
| Berkeley Function Calling Leaderboard (BFCL) | 65.9% | #31 of 47 | Model card | |
| SnakeBench | 16 | #159 of 183 | Official leaderboard | |
| LLM Chess | -637 | #116 of 124 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AraTrust | 84.9% | #9 of 39 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 63.6% | #15 of 27 | Model card |
Every result in the longer documents the lab published with it. 25 of them aren't in the launch post.