LLM Stats (MM-MT-Bench) — leaderboard
Metric: Score (%). Source: llm-stats.com. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Mistral Large 3 | 84.9 |
| 2 | Pixtral Large | 74 |
| 3 | Pixtral-12B | 60.5 |
| 4 | Qwen 3 VL 235B A22B (Thinking) | 8.5 |
| 5 | Qwen 3 VL 235B A22B Instruct | 8.5 |
| 6 | Ministral-3-14B-Instruct-2512 | 8.49 |
| 7 | Qwen 3 VL 32B Instruct | 8.4 |
| 8 | Qwen 3 VL 32B (Thinking) | 8.3 |
| 9 | Qwen 3 VL 30B A3B Instruct | 8.1 |
| 10 | Ministral-3-8B-Instruct-2512 | 8.08 |
| 11 | Qwen 3 VL 8B (Thinking) | 8 |
| 12 | Qwen 3 VL 30B A3B (Thinking) | 7.9 |
| 13 | Ministral-3-3B-Instruct-2512 | 7.83 |
| 14 | Qwen 3 VL 8B Instruct | 7.7 |
| 15 | Qwen 3 VL 4B (Thinking) | 7.7 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-mm-mt-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.