Qwen 3.5 Plus (2026-02-15): benchmark results
Provider: Alibaba. Released 2026-02-16. Access: API.
Aggregate Rating 1749 ± 17, rank #178 of 10870 rated models, from 61 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EgoArgus (Assistance Decision) - Contradictory | 91.01 | Assistance-decision F1 (%; whether help is required, from th | 100 |
| Swiss-Bench SBP-002 | 38.2 | Correct rate (%) on all 395 items of Swiss-Bench SBP-002 (ex | 100 |
| Swiss-Bench SBP-002 - FINMA | 21.9 | Correct rate (%) on the 178 FINMA financial-market regulatio | 100 |
| Swiss-Bench SBP-002 - Hallucination Detection | 30.2 | Correct rate (%) on the 63 hallucination-detection items (pl | 100 |
| Swiss-Bench SBP-002 - Jurisdiction Discrimination | 69 | Correct rate (%) on the 58 jurisdiction-discrimination items | 100 |
| Swiss-Bench SBP-002 - Legal-CH | 63.9 | Correct rate (%) on the 169 Swiss federal law items (data pr | 100 |
| Swiss-Bench SBP-002 - Statutory Interpretation | 30.4 | Correct rate (%) on the 46 statutory-interpretation items of | 100 |
| FINESSE-Bench - CFTe-like Level 1 | 86.81 | Accuracy (%) on the 781 CFTe-like Level 1 multiple-choice qu | 98.3 |
| FINESSE-Bench - Trading and Technical Analysis | 84.75 | Accuracy (%) over the trading and technical-analysis group o | 96.7 |
| FINESSE-Bench - VLigaBench-ru | 84.88 | Accuracy (%) on the 324 VLigaBench-ru Russian-language olymp | 96.7 |
| FINESSE-Bench - CMT-like Level 2 | 89.24 | Accuracy (%) on the 251 CMT-like Level 2 multiple-choice que | 95 |
| Swiss-Bench SBP-002 - Gap Analysis | 16.9 | Correct rate (%) on the 59 regulatory gap-analysis items of | 94.4 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-plus-2026-02-15 · How It Works · Data refreshed daily, snapshot 2026-11-12.