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

Long Video Understanding on LongVideoBench (Accuracy)

67Accuracy

Penguin-VL

Updated 6d ago

Evaluation Results

Method	Links
Penguin-VL 2026.03		67
GPT-4o 2026.04		66.7
GPT-4o 2026.04		66.7
GPT-4o 2026.05		66.7
Eagle2.5-8B 2026.04		66.4
VideoChat-Flash-7B 2026.04		64.7
Gemini-1.5-Pro 2026.04		64
Qwen3-VL 2026.05		63.6
NEO-ov 2026.05		63.5
RLER 2026.04		63
Qwen3-VL 2026.03		62.6
InternVL3-9B 2026.04		62.5
InternVL-3.5 2026.03		62.1
InternVL3.5 2026.05		62.1
Video-Zero 2026.05		61.2
Video-Zero 2026.05		61
Video-Zero 2026.05		60.9
VisPlay 2026.05		60.7
Video-Zero 2026.05		60.7
V-Zero 2026.05		60.6
Video-Zero 2026.05		60.6
Video-Zero (Peak) 2026.05		60.6
VisPlay 2026.05		60.5
VisPlay 2026.05		60.5
VisPlay 2026.05		60.4
V-Zero 2026.05		60.4
Video-Zero 2026.05		60.4
Video-Zero 2026.05		60.4
Video-Zero (Iter 4) 2026.05		60.4
Video-Zero 2026.05		60.4
V-Zero 2026.05		60.3
InternVL3.5-4B 2026.05		60.3
Video-Zero 2026.05		60.2
Video-Zero 2026.05		60.2
V-Zero 2026.05		60.1
V-Zero 2026.05		60.1
VisPlay 2026.05		60
mPLUG-Owl3-8B 2026.04		59.8
VideoLLaMA3-7B 2026.05		59.8
Base Model 2026.05		59.8
VisPlay 2026.05		59.8
V-Zero 2026.05		59.8
Base Model 2026.05		59.8
VideoLLaMA3 2026.05		59.8
XComp 2026.04		59.7
VisPlay 2026.05		59.6
V-Zero 2026.05		59.6
Video-Zero 2026.05		59.6
Video-Zero (Peak) 2026.05		59.6
LLaVA-Video-7B 2026.04		59.5
V-Zero 2026.05		59.5
V-Zero 2026.05		59.5
Video-Zero 2026.05		59.4
VisPlay 2026.05		59.3
VisPlay 2026.05		59.2
V-Zero 2026.05		59.2
V-Zero 2026.05		59.2
VisPlay 2026.05		59.2
GPT-4V 2026.04		59.1
Video-Zero 2026.05		59.1
Video-Zero 2026.05		59.1
Video-Zero (Iter 4) 2026.05		59.1
VisPlay 2026.05		58.9
V-Zero 2026.05		58.6
VideoChat-Flash-2B 2026.04		58.3
Video-Zero 2026.05		58.2
LLaVA-Video 2026.05		58.2
VisPlay 2026.05		58
Base Model 2026.05		57.7
Base Model 2026.05		57.7
MACF (LLaVA-OV1.5-8B) 2026.05		57.6
InternVL3.5 2026.05		57.4
VideoLLaMA3 2026.05		57.1
MACF (Qwen3-VL-8B) 2026.05		56.8
NEO-ov 2026.05		56.8
Keye1.5-VL-8B 2026.05		56.4
TimeMarker 2026.04		56.3
Qwen2.5-VL-7B 2026.04		56
Qwen3-VL-30B 2026.05		55.9
Qwen3-VL 2026.05		55.6
InternVL3-2B 2026.04		55.4
Kangaroo 2026.04		54.8
Kangaroo-8B 2026.05		54.8
LLaVA-OneVision1.5-8B 2026.05		54.4
MACF (Qwen2.5-VL-7B) 2026.05		52.5
mPLUG-Owl3 2026.04		52.1
Qwen2.5-VL-72B 2026.05		51
Qwen3-VL-8B 2026.05		50.7
LLaVA-Next-Video-34B 2026.05		50.5
LatentMAS 2026.05		48.5
Qwen2.5-VL-7B 2026.05		48.1
InternVL2.5-VL-8B 2026.05		43.4
Qwen2.5VL-3B 2026.04		43.3
ShareGPT4Video-8B 2026.05		41.8
ShareGPT4Video-8B 2026.04		39.7
MapReduce 2026.05		38.2
OpenAI GPT-5 nano 2026.03		38.1