OpenClaw

OpenClaw is an open-source, self-hosted personal AI assistant platform that connects messaging apps to AI agents running on your own hardware — no loss of data control.

• Official site: https://openclaw.ai
• Docs: https://docs.openclaw.ai

Prerequisites

• OpenClaw deployed
• Base URL: https://api.tokenhub.belivestudios.com
• API Key: create one in "API Tokens" on the BeLive Studio Console
• Model name: exactly matching a model ID in the console model list

Setup

Add a custom OpenAI-compatible provider in the OpenClaw config (the Models / Providers section of the config file or admin UI):

{
  "providers": {
    "belive-studios": {
      "type": "openai-compatible",
      "baseURL": "https://api.tokenhub.belivestudios.com",
      "apiKey": "sk-your-belive-studios-key",
      "models": ["gpt-5.6-sol", "claude-fable-5", "gemini-3.7-flash"]
    }
  }
}

Then select a model under the belive-studios provider in your agent config, save and restart the gateway.

FAQ

Agent replies "model not found"? The model ID must exactly match the BeLive Studio console list and be included in the provider's models array.

No replies on a channel? Check the OpenClaw logs to tell channel issues from model-call issues; verify key and network by calling https://api.tokenhub.belivestudios.com/v1/chat/completions with curl.