Quickstart: Send your first request on the hosted gateway
This topic walks you through connecting your first LLM provider or MCP server and sending a test request through the gateway.
Prerequisites
Before you begin, make sure you have the following:
- Access to a Fabric AI Gateway organization, with permission to create connections.
- An API key or token from the provider you want to connect.
- The models that you want to expose through the route.
- Postman, if you want to send the test request from a Postman collection.
Connect a provider
You can connect an LLM provider or an MCP server. This topic uses an LLM provider. To connect an MCP server, see Connect an MCP server.
To connect a provider, do the following:
- In the sidebar, select LLMs (or MCP servers), and then select the Providers tab.
- Find your provider, and then select Connect.
- On the Create LLM connection page, enter your provider API key in Upstream auth, and review the other values.
- Under Gateway auth, keep Gateway key selected, and then select Create key. Choose an Owner, select Create key, and copy the key. The console shows it only once. Creating the connection doesn’t create a gateway key for you. For other ways to create one, see Create a gateway key.
- Select Create connection. You now have a connection that holds your provider credential and a route that your clients can call with the gateway key.
For a description of the connection settings, see Connect an LLM provider.
Send a test request from Postman
Before you point an application at the gateway, test the route to confirm that it works. To send a request to an OpenAI route from Postman, do the following:
-
In the sidebar, select LLMs, and then select the Routes tab. Find your OpenAI route, and then select its Copy LLM base URL icon.
-
In Postman, create a request with the method
POST. For the URL, paste the base URL that you copied, followed by/v1/chat/completions. If the route’s only path ends in/*, the copied URL keeps the/*. Remove it before you add the endpoint path. -
In the Headers tab, add
fg-api-keywith your gateway key as the value. For more information, see Authenticate requests with Key Auth. -
In the Body tab, select raw and JSON, and then enter the following body:
-
Select Send.
The request doesn’t include your provider’s API key, because the gateway adds the key that you saved on the connection.
If you run your own data plane, the copied URL has no host until an admin sets the Base URL in Admin > Data plane.
The route’s ⋯ menu also offers Copy models URL (GET), Copy chat URL (POST), Copy models cURL (GET), and Copy chat cURL (POST), when the route exposes those endpoints.
The response is a standard OpenAI chat completion (200 OK) with the model name and the assistant’s reply.
The response looks like the following, shortened here. Your id, model, and reply text differ:
The response also has an X-Request-Id header. To find the request in Traces, see See your requests. A 200 OK response confirms that the gateway accepted your gateway key and proxied the request to the upstream provider.
If the request returns 401 Unauthorized, see Troubleshoot authentication errors.
To send the request with cURL, or to call an Anthropic route, see Authenticate and send a request.
Next steps
- Learn the platform model in How Fabric works before you expand your deployment.
- To add authentication, rate limiting, routing, or transforms to a route, see Control traffic with plugins.
- To observe traffic through the gateway, see Monitor gateway traffic.
- To give your team access, see Administer your organization.

