Meta

Verified : Code examples on this page have been automatically tested and verified.

Route agentgateway LLM traffic to the Meta Model API.

Configure Meta as an LLM provider in agentgateway. The meta preset supports the Chat Completions, Messages, and Responses APIs.

Before you begin

Install the agentgateway binary.

You need agentgateway 1.6.0 or later and a Meta Model API key. Create a key in the Meta Model API dashboard.

Configuration

Use the meta preset to route requests to the Meta Model API. The preset supplies the provider URL and supported request formats.

You can add the model in the UI or in your configuration file.

To add the model in the UI, follow these steps. The steps create the same model as the Config file tab.

Set your API key in the META_API_KEY environment variable. The UI checks that the variable exists when you save the model.
```
export META_API_KEY='<your-api-key>'
```
Create a configuration file with an empty llm section, and start agentgateway in the same terminal. Without an llm section, the UI does not show the LLM > Models page.
```
cat > config.yaml <<'EOF'
# yaml-language-server: $schema=https://agentgateway.dev/schema/config
llm:
  models: []
EOF
agentgateway -f config.yaml
```
Open LLM > Models, such as http://localhost:15000/ui/llm/models, and click Add model.
From the Provider dropdown, select Meta.
Keep the default meta/* in the Incoming model match field.
For the Provider API key, select Env var, and enter META_API_KEY.
For Outgoing model, keep Strip meta/. This option removes the meta/ prefix before the request is forwarded. To send every request to one model instead, select Explicit, and enter the model name, such as muse-spark-1.3. This option sets params.model.
Click Save model.

The form has no field for params.baseUrl. To override the base URL, click the Raw icon (<>) in the drawer header and add the field, or use the Config file tab.

Whether the UI can save depends on your storage mode. For more information, see Save changes from the UI.

Set your API key in the terminal where you will run agentgateway.
```
export META_API_KEY='<your-api-key>'
```

Create a config.yaml file with the following configuration.

cat > config.yaml <<'EOF'
# yaml-language-server: $schema=https://agentgateway.dev/schema/config
llm:
  models:
  - name: "meta/*"
    provider: meta
    params:
      apiKey: "$META_API_KEY"
      # Optional: use this upstream model for every matching request.
      # model: muse-spark-1.3
      # Optional: override the default provider URL.
      # baseUrl: https://api.meta.ai/v1
    transformation:
      model: llmRequest.model.stripPrefix("meta/")
EOF

Setting	Description
`name`	The model name to match in incoming requests. The `meta/*` pattern matches any model name that starts with `meta/`.
`provider`	The provider preset. Set to `meta`.
`params.apiKey`	Your Meta Model API key. Use `$META_API_KEY` to read the key from the environment.
`params.model`	Optional. The upstream model to use for every matching request. Omit to use the model from the request.
`params.baseUrl`	Optional. Overrides the provider URL. Defaults to `https://api.meta.ai/v1`.
`transformation.model`	Removes the `meta/` prefix from the model name before the request is forwarded, so the provider receives its own model name.

Validate the configuration.

agentgateway -f config.yaml --validate-only

Example output:

Configuration is valid!

Start agentgateway.
```
agentgateway -f config.yaml
```

Example request

From another terminal, send a Chat Completions request to the gateway. This example uses muse-spark-1.3, as shown in the Meta quickstart. Use a model that your Meta account can access.

curl http://localhost:4000/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "meta/muse-spark-1.3",
    "messages": [{"role": "user", "content": "Hello from Meta!"}]
  }'

Hugging Face Mistral AI

Was this page helpful?

Meta

Before you begin

Configuration

Example request

What could be improved?