Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief
BenchLM

Multimodal Multi-disciplinary Video Understanding (MMVU)

A benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, emphasizing temporal reasoning and comprehension over video content.

Data verified 25 confirmed releases in the last 30 daysStart free brief

Benchmark score on MMVU — August 21, 2026

We mirror the published score view for MMVU. Qwen3.8 Max leads the public snapshot at 82.4%, followed by Kimi K2.5 (80.4%) and dots3-note Preview (79.9%). We do not use these results to rank models overall.

1Open

Qwen3.8 Max

Alibaba

82.4%
Overall 78.95Context 1M
2Open

Kimi K2.5

Moonshot AI

80.4%
Overall 58.68Context 256K
3Open

dots3-note Preview

Dots Studio

79.9%
Overall 68.81Context 512K
6 modelsMultimodal & GroundedCurrentDisplay onlyUpdated August 21, 2026

Benchmark score table (6 models)

Score
1
Qwen3.8 MaxAlibaba · Open weight
82.4%
2
Kimi K2.5Moonshot AI · Open weight
80.4%
3
dots3-note PreviewDots Studio · Open weight
79.9%
4
Qwen3.5-122B-A10BAlibaba · Open weight
74.7%
5
Qwen3.5-27BAlibaba · Open weight
73.3%
6
Qwen3.5-35B-A3BAlibaba · Open weight
72.3%

The published MMVU snapshot places Qwen3.8 Max first at 82.4%. The third row is 2.5 points behind. The broader top-10 range is 10.1 points, so the table still separates the published systems.

6 models have been evaluated on MMVU. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. MMVU is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About MMVU

Year

2026

Tasks

Video understanding

Format

Video reasoning benchmark

Difficulty

Multi-disciplinary multimodal video reasoning

MMVU is a useful video-understanding benchmark for BenchLM because it appears in frontier provider tables and complements Video-MME with a more discipline-oriented video reasoning slice.

Kimi K2.5 benchmark release surface

BenchLM freshness & provenance

Version

MMVU 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does MMVU measure?

A benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, emphasizing temporal reasoning and comprehension over video content.

Which model scores highest on MMVU?

Qwen3.8 Max by Alibaba currently leads with a score of 82.4% on MMVU.

How many models are evaluated on MMVU?

6 AI models have been evaluated on MMVU on BenchLM.

Compare Top Models on MMVU

Qwen3.8 Max vs Kimi K2.5Kimi K2.5 vs dots3-note Previewdots3-note Preview vs Qwen3.5-122B-A10BQwen3.5-122B-A10B vs Qwen3.5-27B
Last updated: August 21, 2026 · BenchLM version MMVU 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.