PlayOn

Venice

Privacy-focused OpenAI-compatible API — common default for PlayOn in-app agents.

Models

See Getting started with AI for the full pick list.

  • Minimum: gpt-4o-mini
  • Recommended: grok-4-5 (Home default)
  • Also works: llama-3.3-70b, mistral-small-3-2-24b, qwen 27B, deepseek-v4-flash, kimi-k2-5
  • Skip: Gemma 27B, GLM flash

Get a key

  1. Create an account at venice.ai.
  2. Open the API / keys area in your Venice dashboard and create an inference API key.
  3. Copy the key once — treat it like a password.

Connect in PlayOn

  1. Settings → In-app agents.
  2. Provider: Venice.
  3. Paste the Venice API key.
  4. Keep the default model (grok-4-5) or use gpt-4o-mini if you want the cheaper floor.
  5. Save, then send a chat message to verify.

Lab / contributor installs may also accept PLAYON_VENICE_API_KEY or VENICE_API_KEY in the environment; the Settings UI is the path for Home installs.

Notes

  • Venice is OpenAI-compatible; PlayOn talks to its chat completions endpoint.
  • Prefer models that support tool calling for server install and ops. grok-4-5 is the Home default; gpt-4o-mini is the cheaper floor.
  • For fully offline LANs, use Ollama with qwen2.5 7B instead.