# ToxiGen

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

Tests hate-speech classifiers on machine-generated implicit toxic and benign statements about 13 minority groups.

## Details

- Address: Online, Global
- Website: [github.com/microsoft/ToxiGen](<https://github.com/microsoft/ToxiGen>)
- Tags: Science, Helpful
- Category: Safety
- Status: Active
- Data licence: unknown
- Data access: open
- Grader: exact
- Organization: Microsoft Research
- Released: 2022-03
- Paper: [arxiv.org/abs/2203.09509](<https://arxiv.org/abs/2203.09509>)
- Repository: [github.com/microsoft/ToxiGen](<https://github.com/microsoft/ToxiGen>)
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is ToxiGen?

ToxiGen is at Online, Global.

### How do I contact ToxiGen?

To contact ToxiGen, visit https://github.com/microsoft/ToxiGen.

## 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/toxigen>)
- [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.