# CRUXEval

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

Tests code reasoning by asking models to predict inputs or outputs for short Python functions.

## Details

- Address: Online, Global
- Website: [github.com/facebookresearch/cruxeval](<https://github.com/facebookresearch/cruxeval>)
- Tags: Helpful, Science
- Category: Coding
- Status: Active
- Data licence: MIT
- Data access: open
- Grader: exact
- Organization: Meta AI
- Released: 2024-01
- Paper: [arxiv.org/abs/2401.03065](<https://arxiv.org/abs/2401.03065>)
- Repository: [github.com/facebookresearch/cruxeval](<https://github.com/facebookresearch/cruxeval>)
- Task count: 800
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is CRUXEval?

CRUXEval is at Online, Global.

### How do I contact CRUXEval?

To contact CRUXEval, visit https://github.com/facebookresearch/cruxeval.

## 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/cruxeval>)
- [Coding in Online](<https://aibenchmarkindex.com/global/online/category/coding.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.