> For the complete documentation index, see [llms.txt](https://docs.stackai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stackai.com/technical-reference/evaluations-and-observability/evaluators-scoring-and-pass-criteria.md).

# Evaluators, Scoring, and Pass Criteria

Evaluators define how a result is assessed. Choose criteria that match the task and explain what an acceptable answer must contain.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FSQ1Sgy933krTyK0S4JpP%2Fuploads%2FrOsY0QHPPXyDA0yHAbPJ%2FG082.mp4?alt=media&token=0293bdbe-4b00-4925-89f5-d69bc476fc68>" %}

## Before you begin

Generated results or sample outputs and access to the supported evaluator catalog.

## Scoring concepts

| Term            | Definition                                                                                                                                                                |
| --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Evaluator       | A configured method for assessing a generated result. Its criteria and score scale determine what the score means.                                                        |
| Scoring         | Applying an evaluator to a generated output and its required context. Missing scores and failed scoring are not passing results.                                          |
| Expected output | The reference value used by an evaluator when its method requires one. It represents the desired behavior, not necessarily a verbatim answer for every evaluation method. |
| Validation      | A check against supported structure and constraints. Passing validation does not prove that generated answers meet every quality requirement.                             |

## Configuration options

| Option         | Purpose                                                                  | Behavior                                                                           |
| -------------- | ------------------------------------------------------------------------ | ---------------------------------------------------------------------------------- |
| Pass criterion | The rule or threshold used to classify an assessed result as acceptable. | A pass applies to that criterion, not automatically to every business requirement. |

## Configure an evaluator

1. Select an evaluator and inspect its required configuration.

![Choose the evaluation method and configure its criteria.](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/5TKP5JQyQyS2wo2RKkSN/assets/g082-g082-01-rule.png)

Choose the evaluation method and configure its criteria.

2. Set its criteria, thresholds, and any required expected output.
3. Review individual scores and explanations before interpreting the aggregate result.

The scores reflect the selected criteria for each evaluated result.

## Behavior and limitations

Different evaluators return different scales and meanings. A high score for one criterion does not prove every business requirement was met. Model-based evaluation can incur usage and vary between runs. Previewing an evaluator does not necessarily save a full evaluation result.

## Related resources

[Configure an evaluator](https://docs.stackai.com/guides-and-quickstarts/evaluations-and-observability/configure-an-evaluator.md)

[Evaluation Datasets and Input Mapping](https://docs.stackai.com/technical-reference/evaluations-and-observability/evaluation-datasets-and-input-mapping.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.stackai.com/technical-reference/evaluations-and-observability/evaluators-scoring-and-pass-criteria.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
