Qwen 3 235B Instruct: benchmark results
Provider: Alibaba. Released 2025-04-29.
Aggregate Rating 1653 ± 28, rank #604 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OPT-BENCH | 60 | EG (20 steps) (self-reported) | 75 |
| Korean Canonical Legal Benchmark | 68.6 | Mean with supporting precedents (imported) | 70 |
| Forge | 65.4 | Overall SR (self-reported) | 65 |
| OPT-BENCH-NP | 57 | Expert-gap closure (EG) at 20 optimization steps with histor | 55.6 |
| WildAgtEval | 58.7 | Complexity-Injected API Call Accuracy (imported) | 44.4 |
| Magis-Bench | 3.84 | AVG (self-reported) | 29.5 |
Interactive version: theaggregate.ai/model?slug=qwen-3-235b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.