Video Reasoning on MMVU mc

82.6Score

GPT-5

Updated 1mo ago

Evaluation Results

Method	Links
GPT-5 2026.01		82.6
Gemini-2.5-Pro 2026.01		78.4
GPT-4o 2026.01		75.4
Gemini-1.5-Pro 2026.01		71.2
ProcessThinker (process-only) 2026.04		68.48
ProcessThinker (outcome + process) 2026.04		67.52
ProcessThinker (outcome-only) 2026.04		67.36
Video-KTR 2026.01		66.6
Video-RTS 2026.01		66.4
VIDEO-R1-7B 2026.04		65.92
Video-KTR 2026.01		65.9
TW-GRPO 2026.01		65.8
Video-KTR 2026.01		65.7
QWEN3-VL-8B-INSTRUCT 2026.04		65.6
Ours 2026.06		65.3
PROCESSTHINKER-SFT 2026.04		64.48
VideoChat-R1 2026.06		64.2
Video-R1 2026.01		63.8
Video-R1 2026.06		63.8
Ours(wo-mv) 2026.06		63.8
Qwen2.5-VL-SFT 2026.01		63.5
Qwen2.5-VL-7B-SFT 2026.06		63.5
Video-R1-wo-image 2026.06		62.7
Qwen2.5-VL-SFT 2026.01		61.6
Qwen2.5-VL-SFT 2026.01		61.3
Video-R1-wo-image 2026.06		60.6
Qwen2.5-VL 2026.01		59.2
Qwen2.5-VL-7B 2026.06		59.2
LLaVA-OV 2026.01		49.2
VILA-1.5 2026.01		49.2
LLaVA-OneVision 2026.06		49.2
VideoLLaMA2 2026.06		44.8