PlayOn

Custom OpenAI-compatible

Point PlayOn at any OpenAI-compatible base URL and model.

Use Custom when you run a proxy, gateway, or self-hosted stack that speaks the OpenAI chat completions API (LiteLLM, vLLM with an OpenAI facade, corporate gateway, etc.). Any OpenAI-compatible endpoint that is not in the Settings preset list belongs here too.

What you need

  • Base URL ending in a path PlayOn can call (often …/v1)
  • Model id your gateway expects
  • API key if the gateway requires Authorization: Bearer …

Connect in PlayOn

  1. Settings → In-app agents → Custom (OpenAI-compatible).
  2. Enter the Base URL (editable for this preset).
  3. Enter the Model id.
  4. Paste the API key if required.
  5. Save and test with a tool-using chat.

Notes

  • If your URL matches a known preset (Venice, OpenAI, NVIDIA, …), prefer that preset so defaults stay correct. See Getting started with AI.
  • Tool schemas follow OpenAI function-calling style — gateways that strip tools will not drive PlayOn well.
  • For local Ollama specifically, use the Ollama preset instead of Custom.