Fx
Connect fx to AudioLab Router.
Configure fx
Create an AudioLab workspace key with llm.write and available credits. Use a model enabled for your workspace. Keep existing client permissions and tool approval settings.
Requires the configurable-model-connections preview build. Add a named provider in ~/.fx/settings.json with protocol openai-chat-completions, base_url, and bearer authentication through AUDIOLAB_API_KEY. Select it with fx provider audiolab and check fx status --json before prompting.
{
"providers": {
"audiolab": {
"protocol": "openai-chat-completions",
"base_url": "https://api.tryaudiolab.ai/v1",
"auth": {
"type": "bearer",
"env": "AUDIOLAB_API_KEY"
},
"model_metadata": {
"openai/gpt-6-astra": {
"context_window": 128000,
"max_output_tokens": 4096,
"supports_tool_use": true
}
}
}
},
"models": {
"audiolab": "openai/gpt-6-astra"
}
}fx provider audiolab
fx status --json
fxSupported Configuration
Custom connections are documented for a preview build, not all published fx releases. Confirm your binary includes that feature and that fx status --json shows the AudioLab endpoint before prompting. fx setup configures a Vercel Gateway credential, so it is not the command to use here.
Verify the Connection
Send a short text prompt, then a read-only tool request in a disposable project. Check the model, request record, and usage in Router. Test streaming and a follow-up turn before using a longer workflow. Client-specific tools and permissions remain managed by fx.
The setup command can check authentication and model discovery without generating a response. See the protocol contracts for model and feature limits.