Qwen3.7 Launch - Humanity's Last Exam (with tools) — leaderboard
Metric: Score (%). Source: qwen.ai. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.6 (Thinking) | 54 |
| 2 | Qwen 3.7 Max | 53.5 |
| 3 | Claude Opus 4.6 (Thinking) | 53 |
| 4 | GLM-5.1 (Thinking) | 52.3 |
| 5 | Qwen 3.6 Plus | 50.2 |
| 6 | DeepSeek V4 Pro (Reasoning, Max Effort) | 48.2 |
Interactive version: theaggregate.ai/benchmark?slug=qwen3-7-launch-humanity-s-last-exam-with-tools · How the rankings work · Data refreshed daily, snapshot 2026-07-22.