Video-MME

Models scored

50

Best score

0.8

Metric

Overall (no subtitles)

Video-MME is an AI evaluation tracked by Epoch AI, with 50 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.

Scored models span August 2023 to June 2025. The best score recorded is 0.8, measured as Overall (no subtitles). Models come from ByteDance, Google DeepMind, Alibaba, Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University, OpenAI, NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University, Shanghai AI Lab, Anthropic.

Full record

Score metric: Overall (no subtitles)
Models scored: 50
Best score recorded: 0.8
Earliest model scored: Aug 20, 2023
Most recent model scored: Jun 18, 2025
Third-party reported: Yes

Leaderboard

01	video-SALMONN-2plus	ByteDance	0.8
02	gemini-1.5-pro-001	Google DeepMind	0.75
03	gemini-1.5-pro-001-feb24	Google DeepMind	0.75
04	Qwen2.5-VL-72B-Instruct	Alibaba	0.73
05	InternVL2_5-78B	Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University	0.72
06	gpt-4o-2024-11-20	OpenAI	0.72
07	gpt-4o-2024-08-06	OpenAI	0.72
08	Qwen2-VL-72B-Instruct		0.71
09	LLaVA-Video-72B-Qwen2		0.71
10	gemini-1.5-flash-001	Google DeepMind	0.7
11	LinVT		0.7
12	Aria		0.68
13	ViLAMP-llava-qwen		0.68
14	Oryx-1.5-32B		0.67
15	LLaVA-OneVision 72B		0.66
16	VideoLLAMA3-7B		0.66
17	LLaVA-Video-7B-Qwen2		0.66
18	LLaVA-Video-7B-Qwen2-TPO		0.66
19	VideoChat-Flash-Qwen2-7B_res448		0.65
20	gpt-4o-mini-2024-07-18	OpenAI	0.65
21	ByteVideoLLM-14B		0.65
22	NVILA-8B	NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University	0.64
23	LiveCC-7B-Instruct		0.64
24	MiniCPM-o-2_6		0.64
25	Qwen2-VL-7B-Instruct		0.64
26	VideoLLAMA2-7B		0.62
27	InternVL2-40B	Shanghai AI Lab	0.61
28	MiniCPM-V-2_6		0.61
29	claude-3-5-sonnet-20241022	Anthropic	0.6
30	claude-3-5-sonnet-20240620	Anthropic	0.6
31	gpt-4-1106-vision-preview	OpenAI	0.6
32	mPLUG-Owl3-7B-241101		0.59
33	TimeMarker		0.57
34	VITA-1.5		0.56
35	kangaroo		0.56
36	VITA		0.56
37	Video-XL-7B		0.56
38	Video-CCAM-7B-v1.2		0.53
39	long-llava-qwen2-7b		0.53
40	LongVA-7B		0.53
41	Qwen-VL-Max		0.51
42	InternVL-Chat-V1-5		0.51
43	SliME-Llama3-8B		0.45
44	Chat-Uni-Vi-7B-v1.5 + 100k SG-WV		0.43
45	Qwen-VL-Chat		0.41
46	Chat-UniVi-7B-v1.5		0.41
47	sharegpt4video-8b		0.4
48	Video-LLaVA-7B		0.4
49	video_chat2_mistral		0.4
50	ST-LLM		0.38

More benchmarks

Epoch Capabilities Index MMLU Terminal-Bench GPQA Diamond OTIS Mock AIME 2024-2025 GSM8K ARC-AGI ARC-AGI-2

SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.

← All benchmarks