> For the complete documentation index, see [llms.txt](https://docs.radar.citadel-ai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.radar.citadel-ai.com/en/application-owner-guide/new-application/set-metrics.md).

# Configure Evaluation Metrics for an Application

From the Nodes screen, select the target application and go to the application settings screen.

After navigating, select `Open Settings` from the metric card.

<figure><img src="/files/5Kuhe9mDBclklIDG8VCX" alt="" width="563"><figcaption></figcaption></figure>

Once on the Metrics page, select `Add Metric` in the top right corner to open the Add Metric screen.

<figure><img src="/files/u9z0QUv1rjNzivOkGaka" alt="" width="563"><figcaption></figcaption></figure>

### Add metrics

Metrics can be added in two ways.

#### 1-a. Assign an Evaluation Package

When you select an Evaluation Package, which is a set of metrics, you can assign all metrics included in the package at once.

For details about Evaluation Packages, see [Create Evaluation Packages](/en/application-owner-guide/evaluation-library/evaluation-packages.md).

<figure><img src="/files/HNHloKd9HVpit0gK80k3" alt=""><figcaption></figcaption></figure>

#### 1-b. Add Evaluation Metrics individually

You can also add Evaluation Metrics individually. Select the metric you want to add, then select the `+` button on the far right.

<figure><img src="/files/CfdAvVu3Qs6pNNeklUhT" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
Even after assigning an Evaluation Package, individual Evaluation Metrics can still be added.
{% endhint %}

#### 2. Configure each metric's parameters

For each metric, configure parameters such as `Action`, `Borderline handling`, and `Remote LLM`.

<table><thead><tr><th width="182.8046875">Parameter</th><th>Overview</th></tr></thead><tbody><tr><td>Action</td><td>Select <code>Block</code> to control the AI system's behavior if an issue is detected in the input or output.<br>Select <code>Flag</code> to let the action proceed while marking the issue for later review.</td></tr><tr><td>Border handling</td><td>For borderline results, selecting <code>Pass</code> treats the case as acceptable, while <code>Fail</code> treats it as an issue.<br>Choose <code>Pass</code> to minimize false positives, or <code>Fail</code> if you want to prevent false negatives for this metric.</td></tr><tr><td>Remote LLM</td><td>Select <code>Fast Model</code> to prioritize response time and cost, or <code>High-Accuracy Model</code> to prioritize evaluation precision. This option is disabled if the selected metric is a <code>Local</code> type.</td></tr></tbody></table>

{% hint style="info" %}
Evaluation metrics set to `Block` are evaluated synchronously by Radar, When Block is selected for an evaluation metric, Radar evaluates it synchronously and returns the results to the application upon completion. Since evaluation triggers requests to all configured evaluation models, the overall response time depends on the slowest model.

**To maintain stable latency, we recommend utilizing the same model (Local / Fast / High-Accuracy) for all metrics assigned to `Block`.** This does not apply to metrics set to `Flag`, which are processed asynchronously.
{% endhint %}

#### 3. Apply the changes

After adding, updating, or deleting metrics, select the `Save changes` button at the bottom of the screen to confirm the changes.

<figure><img src="/files/gZxt9iT1wQfUQLiS02SO" alt=""><figcaption></figcaption></figure>