Multimodal Reasoning on MMMU (test) benchmark leaderboard · SOTA2 Research

Multimodal Reasoning on MMMU (test)

64.7AccuracyGPT-4o

Evaluation Results

Method	Links
GPT-4o2026.01		64.7
InternVL2.5-78B2024.12		61.8
Qwen2.5-VL-DP2025.07		59.4
Qwen2.5-VL-7B2025.07		58.6
InternVL2.5-38B2024.12		57.6
InternVL2-Llama3-76B2024.12		55.1
NVLM-D-72B2024.12		54.6
InternVL2.5-26B2024.12		51.8
Full Prec2025.08		50.44
Qwen2-VL-DP2025.07		49.4
InternVL2-40B2024.12		49.3
InternVL2.5-8B2024.12		48.9
Qwen2-VL-7B2025.07		47.8
VILA-1.5-40B2024.12		46.9
Qwen2-VL2026.01		46.6
InternVL2.5-4B2024.12		46.3
LinMU-NV2026.01		44.6
NVILA2026.01		44.4
InternVL2-8B2024.12		44.3
InternVL2-26B2024.12		43.8
LLaVA-OV2026.01		42.8
InternVL22026.01		42.6
InternVL2-4B2024.12		41.4
InternVL-Chat-V1.52024.12		41
InternVL2.5-2B2024.12		38.2
LLaVA-1.52025.07		36.4
InternVL2.5-1B2024.12		35.8
InternVL2-2B2024.12		34.7
VLMQ2025.08		33.67
SPHINX2025.07		32.9
InstructBLIP2025.07		32.9
InternVL2-1B2024.12		32.8
GPTQ2025.08		32.22
EntropyPrune2026.02		30
FastV2026.02		29.3
LLaVA-Next-7B2026.02		28
PDrop2026.02		27.3
CDPruner2026.02		27.3
Frequent Guess2025.07		26.8
DART2026.02		26.7
DivPrune2026.02		25.3
GPTAQ2025.08		22.25
Random Chance2025.07		22.1