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

# Create an evaluation

> From blank page to a runnable eval in two minutes.

# Create an evaluation

An evaluation is a named set of test cases plus the scorers that decide pass or
fail. You will create one, add a couple of cases, and pick a scorer.

## Step 1 — Open evaluations

In the sidebar, click **Evaluations**, then **New evaluation**.

<img src="https://mintcdn.com/evalgate/c7l4ztAi8W3wgmvV/help/images/create-evaluation-list.png?fit=max&auto=format&n=c7l4ztAi8W3wgmvV&q=85&s=e13ba1de7a014ae18f87ddcdd7bdba81" alt="A screenshot of the Evaluations list with the New evaluation button highlighted." width="2400" height="1500" data-path="help/images/create-evaluation-list.png" />

## Step 2 — Name it

Give the evaluation a name that says what it checks, for example
`checkout-assistant-hallucination`. Add a short description. Click **Create**.

<img src="https://mintcdn.com/evalgate/c7l4ztAi8W3wgmvV/help/images/create-evaluation-dialog.png?fit=max&auto=format&n=c7l4ztAi8W3wgmvV&q=85&s=a804c78aad85e5fa959cf6006eae9265" alt="A screenshot of the New evaluation dialog with a name and description." width="2400" height="1500" data-path="help/images/create-evaluation-dialog.png" />

## Step 3 — Pick a model and scorer

On the evaluation page:

1. Under **Model**, choose the model this eval runs against.
2. Under **Scorers**, add at least one. Start with a **deterministic** scorer
   (for example, "output contains the order ID") before reaching for an LLM
   judge.

<Note>
  Model-backed scorers need a [connected provider
  key](../evalgate-101/connect-provider-key).
</Note>

## Step 4 — Add cases

Go to the **Test cases** tab and add your first cases. See
[Add test cases](./add-test-cases).

## Up next

[Run the evaluation](./run-evaluation) and read the report.