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

Home/Benchmarks

Jailbreaking on SafeBench evaluated on OpenAI-o1

34.8FS

o1

Updated 1mo ago

Evaluation Results

Method	Links
o1 2024.11		34.8	15.6	64.4	49.4

SOTA Paper

o1

Jailbreak Large Vision-Language Models Through Multi-Modal Linkage

Dataset

SafeBench

Follow for update

@wizwand_team Discord

Related Benchmarks

Jailbreak Attack on SafeBench Multimodal Jailbreaking on HADES-Dataset Safety Evaluation on HADES-Dataset Multimodal Safety Evaluation on SafeBench Multimodal Safety Evaluation on MM-SafeBench

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs