Qwen 3.5 Plus (Thinking): benchmark results
Qwen 3.5 Plus evaluated with thinking enabled. Provider: Alibaba. Released 2026-02-16. Access: API.
Unified ELO 1649 ± 1, rank #248 of 1761 rated models, from 26 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SecCodeBench | 67.08 | Total Score | 97.1 |
| BenchTable | 72.7 | Total Score (%) | 89.5 |
| Vals AI MedQA | 95.21 | Accuracy (%) | 87.2 |
| Vals AI LegalBench | 85.1 | Accuracy (%) | 84.4 |
| Vals AI CorpFin v2 | 65.31 | Accuracy (%) | 77.4 |
| Vals AI LiveCodeBench | 85.33 | Accuracy (%) | 76.8 |
| Vals AI MMLU-Pro | 87.18 | Accuracy (%) | 75.9 |
| LLM2014 Logic 2026-02 | 53.8 | Median Score | 75.6 |
| Vals AI GPQA | 87.37 | Accuracy (%) | 72.6 |
| LLM2014 Logic 2026-03 | 51.81 | Median Score | 70.7 |
| CLBench | 19.8 | Solving Rate (%) | 68.6 |
| Vals AI AIME | 86.04 | Accuracy (%) | 65.8 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-plus-thinking · How It Works · Data refreshed daily, snapshot 2026-09-05.