Calling public models

Raw HTTP

Point any HTTP client at the standard OpenAI base URL. The request body is identical to a chat completion against a Compute instance — only the model field changes.

1curl https://api.qdiv0.com/v1/chat/completions \
2  -H "Authorization: Bearer $QDIV0_API_KEY" \
3  -H "Content-Type: application/json" \
4  -d '{
5    "model": "public:deepseek-v3.2-european",
6    "messages": [{"role": "user", "content": "Hello"}]
7  }'

OpenAI SDK

The OpenAI Python and Node SDKs work without changes. Pass the public model id as the model parameter.

1from openai import OpenAI
2
3client = OpenAI(
4    base_url="https://api.qdiv0.com/v1",
5    api_key="your-api-key",
6)
7
8resp = client.chat.completions.create(
9    model="public:deepseek-v3.2-european",
10    messages=[{"role": "user", "content": "Hello"}],
11)
12print(resp.choices[0].message.content)

Forwarded features

The public-model router forwards the standard OpenAI request features to the underlying provider. Streaming, function calling, JSON mode, and stop sequences work against every catalog model that supports them on its native provider.

The platform reports token usage in the response, so the cost is auditable per call. Streaming responses aggregate the final usage payload the provider emits at the end of the stream.

See Inference for the full request shape, including how to attach a firewall, enable tools, and stream Server-Sent Events.