Qwen 3.7 Plus: benchmark results
Alibaba's balanced Qwen 3.7 Plus tier. Provider: Alibaba. Released 2026-06-02. Access: API.
Aggregate Rating 1768 ± 5, rank #92 of 1656 rated models, from 580 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AgentS4D | 60.15 | Conditional attack success rate (%, cASR): unsafe runs over | 100 |
| AgentS4D (Claude Code) | 60.8 | Conditional attack success rate (%, cASR): unsafe runs over | 100 |
| AgentS4D (Codex CLI) | 63.88 | Conditional attack success rate (%, cASR): unsafe runs over | 100 |
| AgentS4D (Hermes Agent) | 58.13 | Conditional attack success rate (%, cASR): unsafe runs over | 100 |
| AgentS4D (OpenClaw) | 58.02 | Conditional attack success rate (%, cASR): unsafe runs over | 100 |
| CountQA | 77 | Score (%) | 100 |
| Ego3D-Bench | 5.9 | Score | 100 |
| GroundBench - 64 Vertices | 38.5 | Filled-region IoU (0-100), IoU x 100 (N=64 vertex budget; 1, | 100 |
| LLM Stats (DeepPlanning) | 62.3 | Score (%) | 100 |
| LLM Stats (MLVU) | 87.4 | Score (%) | 100 |
| LLM Stats (MathVision) | 90.3 | Score (%) | 100 |
| LLM Stats (OCRBench_V2) | 67.1 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=qwen-3-7-plus · How It Works · Data refreshed daily, snapshot 2026-10-20.