# AI model evaluation: what it is and how it works
> AI model evaluation checks whether a system meets quality and behaviour criteria on defined tasks.
URL: https://www.zorraquino.com/en/dictionary/ai-model-evaluation

# AI model evaluation

AI model evaluation checks whether a system meets quality and behaviour criteria on defined tasks

What it means, how it is used and where it comes up, with its related concepts.

[The whole glossary](https://www.zorraquino.com/en/dictionary)

## What it is and how it works

It uses test cases, metrics and human review depending on the context. It should include frequent, difficult and failure cases. A good result on a public benchmark does not prove that the system works well with an organisation's data and needs.

Also known as

Evals, AI evaluation

Topics

[Artificial intelligence](https://www.zorraquino.com/en/dictionary/category/inteligencia-artificial)

## Related concepts

- [AI governance](https://www.zorraquino.com/en/dictionary/ai-governance)
- [AI guardrails](https://www.zorraquino.com/en/dictionary/ai-guardrails)
- [Generative AI](https://www.zorraquino.com/en/dictionary/generative-ai)

## References

- [Anthropic: evaluating agents](https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents)