Qwen 3 4B 2507 Instruct: benchmark results
Dedicated direct-answer checkpoint from the July Qwen3 update. Does not emit thinking blocks; post-training targets instruction following, tool use and multilingual knowledge. Provider: Alibaba. Released 2025-08-06. Access: Open.
Aggregate Rating 1510 ± 8, rank #747 of 1635 rated models, from 671 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Industrial-Instruction Panasonic QA | 28.5 | Set-match accuracy (%; share of the 1,000 held-out multi-sel | 100 |
| Kurakurai French LLM - AIME 2025 | 38 | Accuracy (%) | 100 |
| Kurakurai French LLM - MATH-500 | 92.46 | Accuracy (%) | 100 |
| Kurakurai French LLM - MBPP | 76.03 | pass@1 (%) | 100 |
| Kurakurai French LLM - MGSM | 93.08 | Accuracy (%) | 100 |
| Kurakurai French LLM - MultiIF | 71.35 | Accuracy (%) | 100 |
| Plan-RewardBench - Multi-Turn Planning (Easy) | 75 | Pairwise accuracy (%) on the 109 easy multi-turn planning pa | 100 |
| SAHARA - Paraphrase | 36.29 | spBLEU-1K | 100 |
| SciEval (K-12) - Evidence Match Rate | 36.04 | Evidence match rate (%): share of predicted evidence sentenc | 100 |
| SciEval (K-12) - Macro-F1 | 29.5 | Macro-F1 (%) of the predicted criterion score (0 not address | 100 |
| SciEval (K-12) - Quadratic Weighted Kappa | 16.23 | Quadratic weighted kappa (x100, -100 to 100) between the pre | 100 |
| Kurakurai French LLM - AIME 2024 | 53.33 | Accuracy (%) | 97.6 |
Interactive version: theaggregate.ai/model?slug=qwen-3-4b-2507-instruct · How It Works · Data refreshed daily, snapshot 2026-10-15.