Benchmark
MVBench
What is MVBench?
A comprehensive multi-modal video understanding benchmark covering 20 challenging video tasks that require temporal understanding beyond single-frame analysis…
- Released
- 2023-11-28
- Modality
- multimodal
- Evaluates
- multimodal, reasoning, spatial reasoning, video, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 17
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/mvbench?top_n=500
MVBench results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Qwen2-VL-72B-Instruct | Qwen | 0.736 | 2024-08-29 | evidence |
| Qwen2.5 VL 72B Instruct | Qwen | 0.704 | 2025-01-26 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.696 | 2025-01-26 | evidence |
| Qwen2.5-Omni-7B | Qwen | 0.703 | 2025-03-27 | evidence |
| Qwen3 VL 30B A3B Instruct | Qwen | 0.723 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Thinking | Qwen | 0.72 | 2025-09-22 | evidence |
| Qwen3 VL 32B Instruct | Qwen | 0.728 | 2025-09-22 | evidence |
| Qwen3 VL 32B Thinking | Qwen | 0.732 | 2025-09-22 | evidence |
| Qwen3 VL 4B Instruct | Qwen | 0.689 | 2025-09-22 | evidence |
| Qwen3 VL 4B Thinking | Qwen | 0.693 | 2025-09-22 | evidence |
| Qwen3 VL 8B Instruct | Qwen | 0.687 | 2025-09-22 | evidence |
| Qwen3 VL 8B Thinking | Qwen | 0.69 | 2025-09-22 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.766 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.746 | 2026-02-24 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.748 | 2026-02-24 | evidence |
| Qwen3.6-27B | Qwen | 0.755 | 2026-04-21 | evidence |
| Qwen3.6-35B-A3B | Qwen | 0.746 | 2026-04-16 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.