Inference
EN
MMVU benchmark maintainers

Multimodal Multi-disciplinary Video Understanding

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

ModelsModels with a result.8
Top resultBest result on this benchmark.—
Top-3 spreadPoints from first to third.—
YearYear of release.2026

Result and price

5 of 8 results

Results

8 results
#ModelResultThe score from the source.
—Qwen3.8 Max82.4%
—GLM-5.3-Flash80.5%
—Kimi K2.580.4%
—dots3-note Preview79.9%
—Qwen3.5-122B-A10B74.7%
—Qwen3.5-27B73.3%
—Qwen3.5-35B-A3B72.3%
—LFM3-VL-2B52.9%