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

Multi-task Multimodal Understanding on MMT-Bench (val)

72.7Score

GPT-5

Updated 1mo ago

Evaluation Results

Method	Links
GPT-5 2026.04		72.7
Qwen3-Omni 2026.04		70.9
Gemini 2.5 Flash 2026.04		70.7
Qwen3-Omni 2026.04		70.4
Gemini 2.5 Flash 2026.04		70
MiniCPM-o 4.5 2026.04		69.7
MiniCPM-o 4.5 2026.04		69.7
Qwen3-VL 2026.04		68.1
InternVL3.5 2026.04		66.7
Qwen3-VL 2026.04		60.9