# AdvBench

AdvBench is an AI benchmark; this page records what it measures and traces its data, grader, and scores to primary sources.

Measures whether adversarial suffixes induce aligned language models to produce targeted harmful behavior.

## Details

- Address: Online, Global
- Website: [github.com/llm-attacks/llm-attacks](<https://github.com/llm-attacks/llm-attacks>)
- Tags: Helpful, Viral, Science
- Category: Safety
- Status: Active
- Data licence: MIT
- Data access: open
- Grader: exact
- Organization: Zou et al.
- Released: 2023-07
- Paper: [arxiv.org/abs/2307.15043](<https://arxiv.org/abs/2307.15043>)
- Repository: [github.com/llm-attacks/llm-attacks](<https://github.com/llm-attacks/llm-attacks>)
- Task count: 520
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is AdvBench?

AdvBench is at Online, Global.

### How do I contact AdvBench?

To contact AdvBench, visit https://github.com/llm-attacks/llm-attacks.

## Source

From [Benchmark primary sources](https://aibenchmarkindex.com/about), license CC0-1.0, checked 2026-10-15.

Last updated October 15, 2026.

## Links

- This benchmark on the [web page](<https://aibenchmarkindex.com/global/online/advbench>)
- [Safety in Online](<https://aibenchmarkindex.com/global/online/category/safety.md>)
- [More benchmarks in Online](<https://aibenchmarkindex.com/global/online.md>)

> **Check before you compare.** Benchmark methods and leaderboards change. Follow the cited primary source before using a result in a decision.