> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mithunai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Run an answer-quality eval on an assistant

> Ask an assistant up to 10 known questions and score each answer for grounding, citations, expected sources and facts, and correct abstention.

Asks the assistant each case through its normal answer path and scores what comes back by comparing text, never with a model. See [Answer quality evals](https://docs.mithunai.com/assistants/answer-quality.md) for what each check means. Each case counts as one answer against your [monthly allowance](https://docs.mithunai.com/analyze/usage-and-limits.md); no conversation is created and nothing is stored. Requires the owner, admin or editor role.

With no model provider configured, `status` is `cannot_run` and nothing is asked. When the model or your allowance runs out partway, the cases scored before it are returned.

**assistant\_id** `string` *required*

The assistant to test (UUID). Another organization's assistant returns `404`.

**cases** `object[]` *required*

One to 10 cases.

**case**

**question** `string` *required*

A question someone would ask, at most 500 characters.

**expected\_facts** `string[]`

Up to 10 phrases a right answer states.

**expected\_sources** `string[]`

Up to 5 strings found in the title or address of a source it should cite.

**should\_abstain** `boolean` (default: false)

`true` when the right answer is to decline.

## Response

**status** `string` *required*

`done`, or `cannot_run`. Treat a value you do not recognise as `cannot_run`.

**problem** `string | null` *required*

`no_model`, `model_unavailable` or `allowance_reached` when it could not run.

**message** `string` *required*

Why it could not run, in a sentence. Empty when it ran.

**total** `integer` *required*

Cases scored.

**passed** `integer` *required*

Cases that passed.

**pass\_rate** `number | null` *required*

`passed / total`, or `null` with no cases scored.

**results** `object[]` *required*

One per case, in order: `question`, `should_abstain`, `abstained`, `answer`, `sources` (`title`,
`url`, `snippet`), `groundedness` (0 to 1, or `null` when it declined), `grounded`,
`citations_correct`, `source_hit` (`null` when the case names no source), `facts_found`,
`facts_missing`, `abstention_correct` and `passed`.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.