Share your thoughts, 1 month free Claude Pro on usSee more

Multi-discipline Multimodal Reasoning on MMMU

61.3Accuracy

SophiaVL-R1-7B

Updated 25d ago

Evaluation Results

Method	Links
SophiaVL-R1-7B 2025.05		61.3
Qwen2.5-VL-7B-Instruct 2025.05		59.1
Qwen2.5-VL-7B-Instruct 2025.05		58.7
Qwen2.5-VL-7B-Instruct 2025.05		58
LLaVA-OneVision-72B 2025.05		56.8
GPT-4V 2025.05		56.8
InternVL2.5-8B-VisualPRM 2025.05		56.2
R1-Onevision-7B 2025.05		51.6
Cambrian-1-34B 2025.05		49.7
LLaVA-OneVision-7B 2025.05		48.8
URSA-8B 2025.05		43.1
Vanilla 2026.03		36.1
QMoP 2026.03		36.1
LDP-v2 2026.03		35.6
C-Abstractor 2026.03		35.4
QMoP 2026.03		35.1
TokenPacker 2026.03		35
FasterVLM 2026.03		34.9
FasterVLM 2026.03		34.8
MQT-LLaVA 2026.03		34.8
Vanilla 2026.03		34.6
Pixel-Shuffle 2026.03		34.6
FasterVLM 2026.03		34.6
QMoP 2026.03		34.6
MQT-LLaVA 2026.03		34.4
Pixel-Shuffle 2026.03		34.4
TokenPacker 2026.03		34.3
LDP-v2 2026.03		34.2
C-Abstractor 2026.03		34.1
LDP-v2 2026.03		33.9
TokenPacker 2026.03		33.8
C-Abstractor 2026.03		33.4
Pixel-Shuffle 2026.03		32.7