Benchmark

MMT-Bench

MMT-Bench is a comprehensive multimodal benchmark for evaluating Large Vision-Language Models towards multitask AGI. It comprises 31,325 meticulously…

Modality
multimodal
Categories
multimodal, reasoning, general, vision
Openness
unknown
Source
llm_stats
Reported scores
4

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
DeepSeek VL2DeepSeek0.6362024-12-13evidence
DeepSeek VL2 SmallDeepSeek0.6292024-12-13evidence
DeepSeek VL2 TinyDeepSeek0.5322024-12-13evidence
Qwen2.5 VL 7B InstructQwen0.6362025-01-26evidence

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.