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

Video Understanding on Video-MME (Segmented by Length)

78.18Overall Score

HiMu

Updated 24d ago

Evaluation Results

Method
HiMu 2026.03	78.18	80.88	77.19	76.53
HiMu 2026.03	76.11	78.92	75	74.49
Uniform 2026.03	73.81	76.96	73.03	71.43
TWW 2026.03	73.41	83.22	70.11	66.89
HiMu 2026.03	73.22	78.55	71	69.9
HiMu 2026.03	71.35	76.92	70.4	66.5
TWW 2026.03	70	78.89	67.11	64
T* 2026.03	69.77	73.66	67.39	68.12
VideoChat-A1 2026.03	69.7	-	-	-
TWW 2026.03	69.48	79	67.11	62.33
Instruct 2026.03	69.3	78.22	66.78	62.89
Uniform 2026.03	68.95	78.43	67.11	61.22
Thinking 2026.03	68.89	77.78	66.67	62.22
BOLT 2026.03	68.74	69.58	67.87	68.73
Qwen3-VL-8B 2026.03	68.4	79	68.1	58.1
HiMu 2026.03	68.28	71.56	65.7	67.48
AKS 2026.03	67.98	70.05	65.1	68.73
HiMu 2026.03	67.65	71.87	66.99	63.94
Qwen3 + ColorTrigger (Ours, 8B) 2026.03	67.6	77.1	66.6	59
HiMu 2026.03	67.09	73.08	65.1	62.86
Uniform 2026.03	67	71.8	62.2	-
InfoShot 2026.03	67	71.5	62.4	-
Uniform 2026.03	66.63	75.41	67.39	56.55
Video-XL-2 2025.08	66.6	-	-	-
Qwen3 + ColorTrigger (Ours, 8B) 2026.03	66.5	74.8	66.6	58.2
Uniform Sampling 2026.03	66.36	76.34	66.31	55.58
LLaVA-Video w/ Video-EM 2025.08	66.2	77.6	64.4	56.6
InternVL-3.5 + ColorTrigger (Ours, 8B) 2026.03	66.1	76.8	66.3	55.1
InternVL-3.5-8B 2026.03	65.6	76.7	65.3	54.7
VSUMM 2026.03	65.2	67.8	62.5	-
VideoZoomer 2026.03	65.2	-	-	55.8
- 2025.10	64.59	75.78	63.33	54.67
LLaVA-Video 2025.08	64.4	74.6	62.9	55.7
Qwen3-VL-8B grayscale 2026.03	63.9	70.8	64.6	56.2
Uniform 2026.03	63.55	72.26	62.33	54.85
TransNet 2026.03	63.2	65.8	60.4	-
VSLS 2026.03	63	71.9	61.9	55.2
Uniform 2026.03	62.99	73.31	60.29	55.83
InternVL-3.5 + ColorTrigger (Ours, 8B) 2026.03	62.8	71.9	63	53.6
Uniform 2026.03	62.57	72.49	61.01	53.82
GPT-4o 2025.08	62.5	71.4	61	55.2
TimeChat-Online-7B 2026.03	62.4	-	-	48.4
Qwen2.5-VL w/ Video-EM 2025.08	62	72.4	60.3	53.4
SC 2025.10	61.81	72.11	61.78	51.56
Qwen2-VL w/ Video-EM 2025.08	61.2	70.7	58.8	52.3
FLoC 2025.10	61.04	71.68	60.56	50.89
LongVU 2025.08	60.6	64.7	58.2	59.5
TS-LLaVA 2025.10	60.37	71	59.56	50.56
K-means 2025.10	60.37	71.78	59.22	50.11
DivPrune 2025.10	60.3	69.11	59.22	52.56
Qwen2-VL w/AKS 2025.08	59.9	-	-	-
Qwen2-VL w/FOCUS 2025.08	59.7	-	-	-
LongVU 2025.10	59.67	68.89	58.44	51.67
LLaVA-OV w/ Video-EM 2025.08	59.6	70.1	57.2	51.5
Qwen2-VL w/BOLT 2025.08	59.5	69.4	57.5	51.5
FLoC 2025.10	59.26	69.33	57.44	51
Qwen2.5-VL 2025.08	59.2	70.6	56.7	50.6
SC 2025.10	58.85	68.44	56.56	51.56
DivPrune 2025.10	58.44	67.67	57.67	50
K-medoids 2025.10	58.41	68.56	56.89	49.78
K-means 2025.10	58.41	69.22	55.67	50.33
Qwen2-VL w/Q-Frame 2025.08	58.3	69.4	57.1	48.3
TS-LLaVA 2025.10	58.19	67.22	56.78	50.56
Frame Uniform 2025.10	57.63	68.78	54.78	49.33
Frame-Voyager 2025.08	57.5	67.3	56.3	48.9
InternVL-3.5-8B grayscale 2026.03	57.3	63.4	57.9	50.6
Dispider-7B 2026.03	57.2	-	-	-
Qwen2-VL 2025.08	56.9	68.7	53.3	48.8
LLaVA-OV 2025.08	56.5	67.8	54.5	47.2
DivPrune 2025.10	56.41	64	56.22	49
FLoC 2025.10	56.22	66.44	54	48.22
Kangaroo-7B 2026.03	56	66.1	55.3	46.6
Pooling 2025.10	55.96	65.33	53.89	48.67
SC 2025.10	55.81	65	54.56	47.89
K-medoids 2025.10	55.7	65.56	53.67	47.89
TS-LLaVA 2025.10	55.67	64.22	55	47.78
Video-XL 2025.08	55.5	64	53.2	49.2
VideoXL-7B 2026.03	55.5	64	53.2	49.2
LongVU 2025.10	54.48	62.56	53.78	47.11
VideoChat-Online-4B 2026.03	54.4	-	-	47.1
VideoTree 2025.08	54.2	-	-	-
K-means 2025.10	54	65.22	49.56	47.22
K-medoids 2025.10	53.93	63.67	50.67	47.44
Frame Uniform 2025.10	53.74	60.78	51.89	48.56
Video-CCAM-14B 2026.03	53.2	62.2	50.6	46.7
LongVA 2025.08	52.6	61.1	50.4	46.2
LongVA-7B 2026.03	52.6	61.1	50.4	46.2
Pooling 2025.10	51.81	60	50.11	45.33
LongVU 2025.10	50.96	57.33	49.78	45.78
Pooling 2025.10	50.37	57	49.11	45
Frame Uniform 2025.10	48.81	52.56	49.44	44.44
VideoLLaMA2 2025.08	47.9	56	45.4	42.1
VideoChat2 2025.08	43.8	52.8	39.4	39.2
ShareGPT4Video 2025.08	43.6	53.6	39.3	37.9
Video-LLaVA 2025.08	41.6	46.1	40.7	38.1
VideoChat2-7B 2026.03	39.5	48.3	37	33.2