🧰 HarmBench

The items that name it 4

• GLM-5.3 and the spread of advanced cyber capabilities
  articleas tools2026-09-29

  Public benchmark measuring how often a model complies with clearly harmful requests

• Understanding the Impact of LLM Watermarking on AI Agent Behavior
  writeupas tools2026-09-26

  Set of 200 harmful behaviors used to evaluate refusal behavior under watermarking

• anmolksachan/AI-ML-Free-Resources-for-Security-and-Prompt-Injection
  repoas tools2026-05-07

  Red teaming methodology/benchmark

• knowledgator/opir-multitask-large-v1.0 · Hugging Face
  toolas classification.key entities