Qwen 3 235B A22B 2507 Instruct FP8: benchmark results
Provider: Alibaba. Released 2025-07-21. Access: Open.
Aggregate Rating 1687 ± 20, rank #234 of 1635 rated models, from 157 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| TriggerBench - Negative Clean PM Accuracy (40K Context) | 82.7 | PM accuracy (%) on the 289 negative controls, where the cont | 100 |
| EuroEval Spanish | 66.09 | Average Score (%) | 99.8 |
| EuroEval Italian Common Sense Reasoning | 86.12 | Common Sense Reasoning Average Score (%) | 99.7 |
| EuroEval Lithuanian NLU | 58.89 | NLU Average Score (%) | 99.6 |
| EuroEval French | 73.2 | Average Score (%) | 99.5 |
| EuroEval Danish Common Sense Reasoning | 85 | Common Sense Reasoning Average Score (%) | 99.4 |
| EuroEval English | 78.94 | Average Score (%) | 99.4 |
| EuroEval French Common Sense Reasoning | 85.74 | Common Sense Reasoning Average Score (%) | 99.4 |
| EuroEval German | 66.96 | Average Score (%) | 99.4 |
| EuroEval German Common Sense Reasoning | 86.72 | Common Sense Reasoning Average Score (%) | 99.4 |
| EuroEval English Common Sense Reasoning | 91.73 | Common Sense Reasoning Average Score (%) | 99.1 |
| EuroEval Lithuanian | 63.7 | Average Score (%) | 99.1 |
Interactive version: theaggregate.ai/model?slug=qwen-3-235b-a22b-2507-instruct-fp8 · How It Works · Data refreshed daily, snapshot 2026-11-05.