Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Video Safety Moderation on SafeWatch
Loading...
92.3
F1 Score
OMNIGUARD-3B
45.396
57.573
69.75
81.927
Dec 2, 2025
F1 Score
Accuracy
Updated 3mo ago
Evaluation Results
Method
Method
Links
F1 Score
Accuracy
OMNIGUARD-3B
Size=3B
2025.12
92.3
82
OMNIGUARD-7B
Size=7B
2025.12
90.9
85.7
GPT-4o
Size=-
2025.12
84.2
77.5
Qwen3-VL-235B
Size=235B
2025.12
79.5
71.8
LLaVA-Video-72B
Size=72B
2025.12
78.2
70.7
Qwen2.5-VL-72B
Size=72B
2025.12
72.5
64
Qwen2.5-Omni-7B
Size=7B
2025.12
68.6
76
Qwen2.5-VL-7B
Size=7B
2025.12
49.7
46.2
LLaVA-Video-7B
Size=7B
2025.12
47.2
44.9
Feedback
Search any
task
Search any
task