Benchmark
MMT-Bench
What is MMT-Bench?
MMT-Bench is a comprehensive multimodal benchmark for evaluating Large Vision-Language Models towards multitask AGI. It comprises 31,325 meticulously curated…
- Modality
- multimodal
- Evaluates
- multimodal, reasoning, general, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 4
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/mmt-bench?top_n=500
MMT-Bench results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek VL2 | DeepSeek | 0.636 | 2024-12-13 | evidence |
| DeepSeek VL2 Small | DeepSeek | 0.629 | 2024-12-13 | evidence |
| DeepSeek VL2 Tiny | DeepSeek | 0.532 | 2024-12-13 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.636 | 2025-01-26 | 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.