MLVU-M

Name: MLVU-M Leaderboard — AI Model Scores
Creator: LLM Stats
License: https://llm-stats.com/legal/terms-of-service

Progress Over Time

Interactive timeline showing model performance evolution on MLVU-M

State-of-the-art frontier

Open

Proprietary

MLVU-M Leaderboard

8 models

			Context	Cost
1	Qwen3 VL 32B Instruct Alibaba Cloud / Qwen Team	33B	—	—
2	Qwen3 VL 30B A3B Instruct Alibaba Cloud / Qwen Team	31B	—	—
3	Qwen3 VL 30B A3B Thinking Alibaba Cloud / Qwen Team	31B	—	—
4	Qwen3 VL 8B Instruct Alibaba Cloud / Qwen Team	9B	—	—
5	Qwen3 VL 4B Thinking Alibaba Cloud / Qwen Team	4B	262K	$0.10 / $1.00
6	Qwen3 VL 4B Instruct Alibaba Cloud / Qwen Team	4B	262K	$0.10 / $0.60
7	Qwen3 VL 8B Thinking Alibaba Cloud / Qwen Team	9B	—	—
8	Qwen2.5 VL 72B Instruct Alibaba Cloud / Qwen Team	72B	—	—

Notice missing or incorrect data?

About this benchmark

What is MLVU-M?

MLVU-M benchmark

MLVU-M is a text benchmark evaluating models on general tasks. LLM Stats tracks 8 models on this benchmark, scored on a 0–1 scale. The current average is 0.8, with the leader at 0.8.

Compare leaders on the best AI for general leaderboards.

Current leaders

Qwen3 VL 32B Instruct from Alibaba Cloud / Qwen Team currently leads the MLVU-M leaderboard with a score of 0.821 across 8 evaluated AI models.

Qwen3 VL 32B InstructAlibaba Cloud / Qwen Team82.1%

Qwen3 VL 30B A3B InstructAlibaba Cloud / Qwen Team81.3%

Qwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team78.9%

FAQ

Common questions about the MLVU-M benchmark and leaderboard.

What is the MLVU-M benchmark?

MLVU-M benchmark

What is the MLVU-M leaderboard?

The MLVU-M leaderboard ranks 8 AI models based on their performance on this benchmark. Currently, Qwen3 VL 32B Instruct by Alibaba Cloud / Qwen Team leads with a score of 0.821. The average score across all models is 0.776.

What is the highest MLVU-M score?

The highest MLVU-M score is 0.821, achieved by Qwen3 VL 32B Instruct from Alibaba Cloud / Qwen Team.

How many models are evaluated on MLVU-M?

8 models have been evaluated on the MLVU-M benchmark, with 0 verified results and 8 self-reported results.

What categories does MLVU-M cover?

MLVU-M is categorized under general. The benchmark evaluates text models.

What is the best open-source model on MLVU-M?

Qwen3 VL 32B Instruct by Alibaba Cloud / Qwen Team is the top-ranked open-source model on MLVU-M, with a score of 0.821 (rank #1).

Which model offers the best value on MLVU-M?

Among models scoring within 10% of the leader, Qwen3 VL 4B Thinking from Alibaba Cloud / Qwen Team is the cheapest, at $0.10 per million input tokens with a score of 0.757.

How recent are the MLVU-M leaderboard results?

The MLVU-M leaderboard was last updated in August 2026 and currently includes 8 evaluated models.