Our new X account is live! Follow @wizwand_team for updates

Multimodal Benchmarking on MM-Bench 37

71.5Accuracy

Mipha-3B+ (Ours)

Updated 3d ago

Evaluation Results

Method	Links
Mipha-3B+ (Ours) 2024.07		71.5
Mipha-3B 2024.07		69.7
Bunny-3B 2024.07		68.6
LLaVA-1.5+ (Ours) 2024.07		67.6
LAF-7B 2024.07		67.3
TinyLLaVA 2024.07		66.9
Imp-v1 2024.07		66.5
mPLUG-Owl2 2024.07		64.5
LLaVA-1.5 2024.07		64.3
Mobile VLM-v2-3B 2024.07		63.2
Qwen-VL-Chat 2024.07		60.6
LLaVA-Phi 2024.07		59.8
Mobile VLM-3B 2024.07		59.6
Shikra 2024.07		58.8
Mobile VLM-v2-1.7B 2024.07		57.7
IDEFICS-80B 2024.07		54.5
Mobile VLM-1.7B 2024.07		53.2
IDEFICS-9B 2024.07		48.2
InstructBLIP 2024.07		36