# MIRACL

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

Evaluates monolingual information retrieval across 18 languages using native-speaker relevance judgments.

## Details

- Address: Online, Global
- Website: [project-miracl.github.io/](<https://project-miracl.github.io/>)
- Tags: Helpful, Science
- Category: Retrieval
- Status: Active
- Data licence: Apache-2.0
- Data access: open
- Grader: exact
- Organization: MIRACL team
- Released: 2022-09
- Paper: [arxiv.org/abs/2210.09984](<https://arxiv.org/abs/2210.09984>)
- Leaderboard: [eval.ai/web/challenges/challenge-page/1881/leaderboard](<https://eval.ai/web/challenges/challenge-page/1881/leaderboard>)
- Repository: [github.com/project-miracl/miracl](<https://github.com/project-miracl/miracl>)
- Task count: 18
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is MIRACL?

MIRACL is at Online, Global.

### How do I contact MIRACL?

To contact MIRACL, visit https://project-miracl.github.io/.

## 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/miracl>)
- [Retrieval in Online](<https://aibenchmarkindex.com/global/online/category/retrieval.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.