LLM Stats (HMMT25): leaderboard
Metric: Score (%). Source: llm-stats.com. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4 (High) | 96.7 |
| 2 | Qwen 3.6 Plus | 94.6 |
| 3 | Qwen 3.5 397B A17B | 92.7 |
| 4 | Qwen 3.6 27B | 90.7 |
| 5 | Qwen 3.5 122B A10B | 90.3 |
| 6 | Grok 4 | 90 |
| 7 | Qwen 3.5 27B | 89.8 |
| 8 | Qwen 3.5 35B A3B | 89.2 |
| 9 | Granite 4.2 30B | 89.17 |
| 10 | Qwen 3.6 35B A3B | 89.1 |
| 11 | Sarvam 105B | 85.8 |
| 12 | Qwen 3 235B A22B 2507 (Thinking) | 83.9 |
| 13 | Qwen 3.5 9B | 82.9 |
| 14 | Granite 4.2 8B | 78.33 |
| 15 | Qwen 3 VL 235B A22B (Thinking) | 77.4 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-hmmt25 · How It Works · Data refreshed daily, snapshot 2026-09-05.