Qwen 3.6 Plus (Non-reasoning): benchmark results
Provider: Alibaba. Released 2026-04-02. Access: API.
Unified ELO 1649 ± 1, rank #350 of 3078 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Natural Intelligence | 42.3 | NatInt Score | 87.2 |
| UGI - Writing | 46.1 | Writing Score | 86 |
| Korean CSAT 2026 (Easy Mode) - Chemistry I | 45 | Points (out of 50) | 63.3 |
| o11y-bench - Pass^3 | 55.56 | Tasks passed on all three attempts, Pass^3 (%) | 57.8 |
| o11y-bench - Pass@3 | 79.37 | Tasks passed on at least one of three attempts, Pass@3 (%) | 53.9 |
| Korean CSAT 2026 (Easy Mode) - Society and Culture | 37 | Points (out of 50) | 51.4 |
| Korean CSAT 2026 (Easy Mode) - Korean | 91.5 | Points (out of 100) | 50 |
| FrontierMath - Tiers 1-3 (v2) | 32.28 | Accuracy (%, 285 private v2 problems) | 48.4 |
| Korean CSAT 2026 (Easy Mode) - Physics I | 29 | Points (out of 50) | 47.3 |
| Korean CSAT 2026 (Easy Mode) - Total | 393.5 | Points (out of 450) | 46.9 |
| Korean CSAT 2026 (Easy Mode) - Life Science I | 35 | Points (out of 50) | 42.9 |
| Korean CSAT 2026 (Easy Mode) - Mathematics | 92 | Points (out of 100) | 38.8 |
Interactive version: theaggregate.ai/model?slug=qwen-3-6-plus-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.