Wikibench

VideoMME w/o sub.

multimodal

Video-MME is a comprehensive evaluation benchmark for multi-modal large language models in video analysis. It features 900 videos across 6 primary visual domains with 30 subfields, ranging from 11 seconds to 1 hour in duration, with 2,700 question-answer pairs. The benchmark evaluates MLLMs' capabilities in processing sequential visual data and multi-modal content including video frames, subtitles, and audio.

Leaderboard

Showing 10 of 10 results

  1. Qwen3.5-122B-A10B
    83.9%
    i
  2. Qwen3.5-27B
    82.8%
    i
  3. Qwen3.5-35B-A3B
    82.5%
    i
  4. Qwen3.6-35B-A3B
    82.5%
    i
  5. Qwen3 VL 235B A22B Instruct
    79.2%
    i
  6. Qwen3 VL 235B A22B Thinking
    79.0%
    i
  7. Qwen3 VL 32B Thinking
    77.3%
    i
  8. Qwen2.5 VL 72B Instruct
    73.3%
    i
  9. Qwen2.5 VL 32B Instruct
    70.5%
    i
  10. Qwen2.5 VL 7B Instruct
    65.1%
    i