MMMU Benchmark Results & Scores | Benchmark Registry

Versions of Massive Multi-discipline Multimodal Understanding
Version	Released	Models	Results
MMMULatest	November 27, 2023	9	9

Latest version: MMMU

Recently reported results for MMMU
Model	Score	Source
Qwen3.6-35B-A3B	81.7%	alibabacloud.com for Qwen3.6-35B-A3B on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab)MMMU AccuracyReported April 17, 2026Launch postAlibaba Cloud CommunitySelf-reportedQwen TeamRecord for Qwen3.6-35B-A3B on Massive Multi-discipline Multimodal Understanding MMMU
Qwen3.5-397B-A17B	85.0%	alibabacloud.com for Qwen3.5-397B-A17B on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab)MMMU AccuracyReported February 17, 2026Launch postAlibaba Cloud CommunitySelf-reportedQwen TeamRecord for Qwen3.5-397B-A17B on Massive Multi-discipline Multimodal Understanding MMMU
Gemini 2.0 Flash	69.3%	storage.googleapis.com for Gemini 2.0 Flash on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab)MMMU AccuracyReported July 2, 2025Technical reportGoogleSelf-reportedGoogle DeepMindRecord for Gemini 2.0 Flash on Massive Multi-discipline Multimodal Understanding MMMU
Gemini 2.5 Pro	82.0%	storage.googleapis.com for Gemini 2.5 Pro on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab)MMMU AccuracyReported July 2, 2025Technical reportGoogleSelf-reportedGoogle DeepMindRecord for Gemini 2.5 Pro on Massive Multi-discipline Multimodal Understanding MMMU
Gemini 2.5 Flash	79.7%	storage.googleapis.com for Gemini 2.5 Flash on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab)MMMU AccuracyReported July 2, 2025Technical reportGoogleSelf-reportedGoogle DeepMindRecord for Gemini 2.5 Flash on Massive Multi-discipline Multimodal Understanding MMMU