Radical Commons

Donated tokens & split-model GPU pools behind one endpoint. Compute, shared.

—

tokens remaining across all donors

—

live pools

—

requests served

—

tokens spent

Use it

curl https://commons.theradicalparty.com/v1/chat/completions \
  -H "content-type: application/json" \
  -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"hi"}]}'

Donate compute

Contribute an API key + budget, or point a self-hosted split-model pool (exo / mesh-llm / ollama) at us:

curl https://commons.theradicalparty.com/donate \
  -H "content-type: application/json" \
  -d '{"name":"my-4090","kind":"endpoint",
       "base_url":"https://my-node.example/v1",
       "models":["qwen3","deepseek-chat"],"tokens":5000000}'

OpenAI-compatible · weighted round-robin cascade · ◉ auto-refresh