5. Make your first inference call

The API is OpenAI-compatible. Point your existing tooling at Pzero, use your Bearer key, and pick a model from the catalog.

curl

curl -sS -X POST "${PUBLIC_URL}/v1/chat/completions" \
  -H "Authorization: Bearer pzero_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.3-70b",
    "stream": false,
    "messages": [{ "role": "user", "content": "Hello from Pzero" }]
  }'

OpenAI SDK

from openai import OpenAI

client = OpenAI(base_url=f"{PUBLIC_URL}/v1", api_key="pzero_YOUR_KEY")

resp = client.chat.completions.create(
    model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Hello from Pzero"}],
)
print(resp.choices[0].message.content)

What happens behind the request

The router selects the cheapest eligible posted offer at or below your max price, forwards once, meters the capacity burn, and settles in USDC with the 10% platform fee.

Notes

• Use a model id from the catalog (GET /v1/models).
• stream: true is not supported in v1.
• A success response includes an X-Balance-Remaining header.
• See Errors & limits for 402, 429, and 503 handling.

Next

• 6. Read your usage