Run OpenClaw on the fastest inference hardware on Earth.

Create an account at app.generalcompute.com and grab your API key. You get $5 in free credit. Or hand this prompt to your agent:

Sign me up for a General Compute API account and get an API key. Fetch instructions from https://docs.generalcompute.com/agent-signup and follow them.

The agent signup flow sends a verification code to your email, then returns an API key. Save it immediately — it's only shown once.

Check what's available. No auth required:

curl https://api.generalcompute.com/v1/public/models

Full pricing and specs at docs.generalcompute.com/models

Copy the prompt below and hand it to your OpenClaw agent, or manually add the provider to your config:

Add General Compute as a custom provider to my OpenClaw config:

Provider name: generalcompute
Base URL: https://api.generalcompute.com/v1
API key: <paste your GC API key>
API type: openai-completions

Then set my default model to one of these General Compute models:
- deepseek-v3.2 (best reasoning, 8k context)
- minimax-m2.5 (best general-purpose, 160k context)
- gpt-oss-120b (fast + cheap, 128k context)

Run a quick test to confirm everything is connected:

curl https://api.generalcompute.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v3.2",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

Replace YOUR_API_KEY with the key from step 1. You should see streamed tokens immediately.