AI & ML ecosystem

Benchmark

MMMU

Massive Multi-discipline Multimodal Understanding — expert-level vision+text.

Category: Vision
Creator: Yue et al.

Links

Official site Paper

Related model families

GPTby OpenAI Claudeby Anthropic Geminiby Google DeepMind Llamaby Meta AI Qwenby Alibaba (Qwen)Grokby xAI

Evaluation data

MMMUMultimodal evaluation