# evals.do

> Evaluate AI model performance automatically with 100+ built-in metrics. Run quality checks, compare models, detect regressions, and optimize prompts. Ship reliable AI products with confidence.

Continuous AI Evaluation on Every Change. Test language models automatically against benchmark datasets. Compare models side-by-side and detect prompt regressions before production.

evals.do is one of the primitives of Platform.do, in the Insight layer (measure, test and trace). Every .do shares one grammar, Business-as-Code, and one SDK.

## Page

- [evals.do](https://evals.do/): Evals.do - Automate AI Model Evaluation | Continuous Testing for LLMs ([Markdown](https://evals.do/index.md))

## Business-as-Code

- [What is Business-as-Code?](https://platform.do/business-as-code): the grammar every .do shares ([Markdown](https://platform.do/business-as-code.md))
- [Platform.do](https://platform.do/): infrastructure for autonomous business ([llms.txt](https://platform.do/llms.txt))

## Docs

- [Documentation](https://docs.platform.do): guides and reference for the whole platform
- [SDK](https://docs.platform.do/sdk): TypeScript SDK, `import { on, send, db, $ } from 'sdk.do'`
- [CLI](https://docs.platform.do/cli): the `do` command
- [MCP](https://docs.platform.do/mcp): one MCP tool, `do`, that runs a script
- [API](https://docs.platform.do/api): REST API

## Other Insight primitives

- [analytics.do](https://analytics.do)
- [experiments.do](https://experiments.do)
- [benchmarks.do](https://benchmarks.do)
- [models.do](https://models.do)

## Optional

- [Full text](https://evals.do/llms-full.txt): this index plus the page in full