AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
Model reports
MVbench
Video understanding benchmark. For non-video-native models, uses 1 fps frame extraction. Scores depend on frame sampling strategy and context length.
Evaluation context and evidence
- Source
- Model reports
- Publisher
- Not recorded
- Benchmark release
- 2025-06-01
- Collected
- Not recorded
- Modality
- Not recorded
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 1
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| GLM-5.3-Flash | 77.8 | percent | temp 1.0, top_p 0.95, max context 256K, 1 fps frame-extraction for non-video-native models | 2026-08-27 · document_publication | GLM-5.3-Flash |