Pi and Oh My Pi

Pi is a terminal coding agent. It supports custom OpenAI-compatible providers through a local config file. Oh My Pi (omp) is a coding-focused fork and works the same way.

Pi

Add a provider to ~/.pi/agent/models.json (create the file if it doesn't exist; merge into it if it does):

~/.pi/agent/models.json

{
  "providers": {
    "standardcompute": {
      "baseUrl": "https://api.stdcmpt.com/v1",
      "api": "openai-completions",
      "apiKey": "$STANDARDCOMPUTE_API_KEY",
      "models": [{ "id": "standardcompute" }]
    }
  }
}

Set the key in your shell, then select standardcompute with /model in Pi:

export STANDARDCOMPUTE_API_KEY=YOUR_API_KEY

If the provider doesn't appear in /model, check the file location, the provider ID and that the environment variable is set in the shell running Pi. See Pi's model configuration docs.

Oh My Pi

Open ~/.omp/agent/models.yml (create it if it doesn't exist). Declare a provider that uses the openai-completions API, with base URL https://api.stdcmpt.com/v1 and your API key, and add standardcompute as its model. Then select it in omp.