Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

LLM Stats (MM-MT-Bench) — leaderboard

Metric: Score (%). Source: llm-stats.com. 17 models tracked.

Top models

#ModelScore
1Mistral Large 384.9
2Pixtral Large74
3Pixtral-12B60.5
4Qwen 3 VL 235B A22B (Thinking)8.5
5Qwen 3 VL 235B A22B Instruct8.5
6Ministral-3-14B-Instruct-25128.49
7Qwen 3 VL 32B Instruct8.4
8Qwen 3 VL 32B (Thinking)8.3
9Qwen 3 VL 30B A3B Instruct8.1
10Ministral-3-8B-Instruct-25128.08
11Qwen 3 VL 8B (Thinking)8
12Qwen 3 VL 30B A3B (Thinking)7.9
13Ministral-3-3B-Instruct-25127.83
14Qwen 3 VL 8B Instruct7.7
15Qwen 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.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.