Hold LLM red-teaming Offense stale
llm-attacks (GCG)
llm-attacks is the research codebase for the GCG (Greedy Coordinate Gradient) method, which generates adversarial suffixes that get aligned open-weight LLMs such as Vicuna and LLaMA-2-Chat to comply with harmful requests, with transfer to other models. It is aimed at ML researchers reproducing AdvBench experiments.
Unmaintained, superseded or risky; avoid new adoption.
Facts
- Repository
- llm-attacks/llm-attacks
- Homepage
- github.com
- Listed via
- Curated
- First seen
- 2026-11-08
- License
- MIT
- Stars
- 4,871
- Last commit
- 2024-08-02
- Checked
- 2026-11-22
Techniques
Listing is not endorsement. Summaries and notes are AI-assisted and reviewed; repository facts refresh automatically.