LLM Stats (HMMT Feb 26): leaderboard
Metric: Score (%). Source: llm-stats.com. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.7 Max | 97.1 |
| 2 | Claude Opus 4.6 (Thinking) | 96.2 |
| 3 | DeepSeek V4 Pro (Max) | 95.2 |
| 4 | DeepSeek V4 Pro (Reasoning, Max Effort) | 95.2 |
| 5 | DeepSeek V4 Flash (Max) | 94.8 |
| 6 | Qwen 3.7 Plus | 92.9 |
| 7 | Kimi K2.6 (Thinking) | 92.7 |
| 8 | GLM-5.2 | 92.5 |
| 9 | DeepSeek V4 Flash (0423) | 91.9 |
| 10 | GLM-5.1 (Thinking) | 89.4 |
| 11 | Qwen 3.6 Plus | 87.8 |
| 12 | Qwen 3.6 27B | 84.3 |
| 13 | Qwen 3.6 35B A3B | 83.6 |
| 14 | GLM-5.1 | 82.6 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-hmmt-feb-26 · How It Works · Data refreshed daily, snapshot 2026-09-05.