Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Video Understanding on TVBench
Loading...
71.5
Accuracy
Seed1.8
44.2416
51.3183
58.395
65.4717
Dec 5, 2025
Jan 8, 2026
Feb 11, 2026
Mar 17, 2026
Apr 20, 2026
May 24, 2026
Jun 28, 2026
Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Accuracy
Seed1.8
2026.03
71.5
Gemini-3-Pro
2026.03
71.1
Gemini 2.5 Pro
2026.03
67.4
Seed-1.5-VL
2026.03
66.6
Gemini 2.5 Pro
Model Class=Closed-sou...
2026.06
59.9
Gemini 3 Pro
Model Class=Closed-sou...
2026.06
56.8
Molmo2-8B + MotionAtlas-Data
Model Class=Open-sourc...
2026.06
54.8
Molmo2-8B
Model Class=Open-sourc...
2026.06
54.1
GPT-5.2
Model Class=Closed-sou...
2026.06
53.8
Qwen3-VL-8B + MotionAtlas-Data
Model Class=Open-sourc...
2026.06
52.9
Molmo2-4B + MotionAtlas-Data
Model Class=Open-sourc...
2026.06
52.1
Qwen3-VL-8B
Model Class=Open-sourc...
2026.06
51.4
Qwen3-VL-4B + MotionAtlas-Data
Model Class=Open-sourc...
2026.06
51.2
Molmo2-4B
Model Class=Open-sourc...
2026.06
50.6
Qwen3-VL-4B
Model Class=Open-sourc...
2026.06
47.2
MECoT
2025.12
45.42
LV-OV
2025.12
45.29
Feedback
Search any
task
Search any
task