Skip to main content
These examples use the direct API host, https://api.nano-gpt.com. It has no IPv6 address, so clients with only IPv6 should use https://nano-gpt.com with the same paths. See API Hosts.

Quick Setup Guide

  1. Install OpenClaw (ClawdBot) and skip any default model setup.
  2. Get your API key from nano-gpt.com/api.
  3. Open your config (Config -> Models -> Raw, or edit ~/.clawdbot/clawdbot.json).
  4. Add a NanoGPT provider block like this:
Cost fields below are estimates in USD per million tokens. Refresh them from the models endpoint when configuring your client; Auto prices can change with eligible routes.
  1. Replace YOUR_NANOGPT_API_KEY_HERE with your actual key.
  2. Restart OpenClaw (or restart the gateway if you are using it).

Some Models To Try

These are examples only. Use the model IDs returned by the models endpoint below.
  • anthropic/claude-opus-5.5
  • openai/gpt-6.1-sol
  • google/gemini-flash-latest
  • minimax/minimax-m3
  • moonshotai/kimi-k3

Get The Most Recent Models

If you want filtered lists:
  • https://api.nano-gpt.com/api/subscription/v1/models (subscription-only)
  • https://api.nano-gpt.com/api/paid/v1/models (paid-only)

Verify Your Setup

Once OpenClaw is running, confirm requests are flowing by checking your usage at nano-gpt.com/usage.

Problems?

Email support at support@nano-gpt.com or use the in-app chat.