Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Safety on MMDS-R
Loading...
96.53
F1 Score
LLaVAShield
19.7156
39.6578
59.6
79.5422
Jun 22, 2026
F1 Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
F1 Score
LLaVAShield
Category=Open-Source M...
2026.06
96.53
SingGuard-8B
Category=Our Models
2026.06
89.63
SingGuard-2B
Category=Our Models
2026.06
89.6
SingGuard-4B
Category=Our Models
2026.06
87.29
GPT-5.1
Category=Closed-Source...
2026.06
77
Qwen3-VL-235B
Category=Open-Source M...
2026.06
73.93
LlavaGuard
Category=Open-Source M...
2026.06
69.72
ShieldGemma-2
Category=Open-Source M...
2026.06
68.57
Gemini3-Pro
Category=Closed-Source...
2026.06
54.75
Qwen3-VL-4B
Category=Open-Source M...
2026.06
47.95
SafeGuard-VL
Category=Open-Source M...
2026.06
46.67
Qwen3-VL-8B
Category=Open-Source M...
2026.06
38.55
LlamaGuard4
Category=Open-Source M...
2026.06
35.67
GuardReasoner-VL-7B
Category=Open-Source M...
2026.06
23.29
LlamaGuard3-Vision-11B
Category=Open-Source M...
2026.06
22.67
Feedback
Search any
task
Search any
task