LLM Stats (HMMT 2025) — leaderboard
Metric: Score (%). Source: llm-stats.com. 33 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 Pro | 100 |
| 2 | GPT-5.2 | 99.4 |
| 3 | DeepSeek V3.2 Speciale | 99.2 |
| 4 | Qwen 3.6 Plus | 96.7 |
| 5 | Kimi K2.5 | 95.4 |
| 6 | Qwen 3.5 397B A17B | 94.8 |
| 7 | nemotron-3-super-120B-a12B | 94.73 |
| 8 | GLM-5.2 | 94.4 |
| 9 | GLM-5.1 | 94 |
| 10 | Qwen 3.6 27B | 93.8 |
| 11 | GPT-5 | 93.3 |
| 12 | Grok 4 Fast | 93.3 |
| 13 | Qwen 3.5 27B | 92 |
| 14 | Qwen 3.5 122B A10B | 91.4 |
| 15 | Qwen 3.6 35B A3B | 90.7 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-hmmt-2025 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.