“Through examining 212 Mallas, we uncovered eight backend LLMs used by Mallas, along with 182 prompts that circumvent the protective measures of public LLM APIs.”

• 212 Mallas
• 8 backend LLMs
• 182 prompts circumvent the protective measures of public LLM APIs

tactics

• uncensored LLMs
• jailbreak prompts

blogs [link] [link]