QuestBench — leaderboard
Metric: Pass Rate (%) (self-reported). Source: benchmarklist.com. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 57.58 |
| 2 | Claude Opus 4.7 | 52.94 |
| 3 | GLM-5.1 | 35.14 |
| 4 | DeepSeek V4 Pro | 32.43 |
| 5 | Gemini 3.1 Pro (Preview) | 31.82 |
| 6 | Qwen 3.6 Plus | 30.95 |
| 7 | MiMo-V2.5-Pro | 22.73 |
| 8 | MiniMax-M2.7 | 16.67 |
| 9 | DeepSeek V3.2 | 7.81 |
Interactive version: theaggregate.ai/benchmark?slug=questbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.