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

Multimodal Reasoning on MMBench EN

83Accuracy

Instruct

Updated 25d ago

Evaluation Results

Method	Links
Instruct 2026.03		83
DPO 2026.03		83
DPO 2026.03		83
DPO-Shift 2026.03		83
ACPO 2026.03		83
Instruct 2026.03		82.8
IPO 2026.03		82
IPO 2026.03		82
β-DPO 2026.03		82
SimPO 2026.03		81.5
SimPO 2026.03		81
β-DPO 2026.03		81
DPO-Shift 2026.03		81
ACPO 2026.03		81
LLaVA-1.5-13B + MMFuser 2024.10		69.9
LLaVA-1.5-13B 2024.10		67.7
LLaVA-1.5-7B + MMFuser 2024.10		67.5
LLaVA-1.5-7B 2024.10		64.3
Qwen-VL-Chat 2024.10		60.6
Shikra 2024.10		58.8
IDEFICS-80B 2024.10		54.5
IDEFICS-9B 2024.10		48.2
Qwen-VL 2024.10		38.2
InstructBLIP 2024.10		36