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

# Publishing Deployments

> Publish a dedicated deployment on Fireworks so other users in your account, or the public, can call it from the API and dashboard without direct GPU access.

By default, deployments are private to your account. Publish a deployment to make it queryable by anyone with a Fireworks API key.

## Publishing a deployment

**Step 1: Find the deployed model ID**

```bash theme={null}
firectl deployed-model list --filter 'model="<MODEL_NAME>" AND deployment="<DEPLOYMENT_NAME>"'
```

**Step 2: Publish it**

```bash theme={null}
firectl deployed-model update <DEPLOYED_MODEL_ID> --public
```

**Step 3: Share the model identifier**

Users can query your public deployment using:

```
<DEPLOYMENT_NAME>
```

## Unpublishing a deployment

To make a deployment private again:

```bash theme={null}
firectl deployed-model update <DEPLOYED_MODEL_ID> --public=false
```

## Use cases

Publishing deployments is useful for:

* **Sharing custom models** – Let others use your trained models
* **Testing with teams** – Allow team members to access deployments
* **Community contributions** – Share optimized model configurations

> [!NOTE]
>
> Published deployments are still hosted and controlled by your account. You pay for all inference costs.


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