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

# Step 1 — Metric

> Pick a predefined metric template or write a custom LLM-as-judge prompt.

The first step asks: *what should this metric measure?* You can pick from a curated list of templates, or click **Use custom prompt** to write your own.

<Frame caption="Step 1 — pick a built-in metric or define your own with a custom prompt">
  <img src="https://mintcdn.com/v2galileo/-aQkdd7oOglUYIo1/images/luna-studio/runs/new-run-metric.png?fit=max&auto=format&n=-aQkdd7oOglUYIo1&q=85&s=72c3c73ca673fc1f2366d7ad429c7d02" alt="Metric step" width="1024" height="659" data-path="images/luna-studio/runs/new-run-metric.png" />
</Frame>

## Pick a metric

The **Metric** select is searchable. It includes trainable Galileo LLM-as-judge metrics available to your workspace, plus custom metrics and prompts your org created.

The picker is organized into three groups:

* **Galileo presets** — built-in Galileo scorers that Luna Studio can train.
* **Custom Galileo metrics** — custom metrics already created in Galileo.
* **Saved custom prompts** — prompts previously authored in Luna Studio.

Metrics that exist but are not trainable yet can appear disabled with a "not trainable yet" suffix. Multimodal metrics are filtered out because Luna Studio trains text metrics today.

## Inspect a selected template

Once you pick a template, the form expands to show a read-only **Metric details** panel:

* **Output type** — the metric's return shape (Boolean, Categorical etc.). See [Output types](/luna-studio/ui/core-concepts#metrics).
* **Step** — the trace step the metric runs against (LLM span, Retriever, Agent span, or Trace).
* **Input step** — the input shape Luna Studio expects for training data, such as a single message, input / output pair, full trace, or full session.
* **Prompt** — the LLM-as-judge prompt the template uses, in a read-only textarea.

## Write a custom prompt

For metrics that don't fit a template, click the dropdown's **Use custom prompt** option (with a `+` icon). The form switches into editable mode.

<Frame caption="Step 1 with Custom prompt expanded — write your own LLM-as-judge prompt for the metric">
  <img src="https://mintcdn.com/v2galileo/-aQkdd7oOglUYIo1/images/luna-studio/runs/new-run-metric-custom.png?fit=max&auto=format&n=-aQkdd7oOglUYIo1&q=85&s=be7734d351d111cf818116403a64adce" alt="Custom metric prompt" width="1024" height="659" data-path="images/luna-studio/runs/new-run-metric-custom.png" />
</Frame>

In custom mode, you fill in:

| Field       | Notes                                                                |
| ----------- | -------------------------------------------------------------------- |
| Name        | Optional display name for the custom metric.                         |
| Output type | Pick one of Boolean / Categorical                                    |
| Step        | Pick LLM span / Retriever / Agent span / Trace.                      |
| Input step  | Pick Single message / Input-output pair / Full trace / Full session. |
| Prompt      | The LLM-as-judge prompt \[required].                                 |

<Tip>
  **Pro tip:** For best results, we recommend first creating your metric in the Galileo console and using the **Autotune** feature to test and refine it on a labelled test dataset. This helps you optimize the metric’s performance before
  launching a full training run in Luna Studio.
</Tip>

## Where to go next

<CardGroup cols={2}>
  <Card title="Step 2 — Test set" icon="database" href="/luna-studio/ui/runs/new-run/step-2-test-set">
    Pick the labelled dataset Luna evaluates against.
  </Card>

  <Card title="Custom metrics" icon="code" href="/luna-studio/ui/metrics/custom-metrics">
    Reference for output types, steps, and prompt-writing tips.
  </Card>
</CardGroup>