API reference

Chat Completions

OpenAI's Chat Completions API, for every model your company can use.

POSThttps://sov.solibra.ai/v1/chat/completions

Request

The body is OpenAI's Chat Completions request: model, messages, tools, response_format, stream and the rest. Send your Sovereign key as Authorization: Bearer.

curl https://sov.solibra.ai/v1/chat/completions \
  -H "Authorization: Bearer $SOV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6.1-sol",
    "messages": [{"role": "user", "content": "What is a gateway?"}]
  }'

reply = client.chat.completions.create(model="gpt-6.1-sol", messages=[{"role": "user", "content": "What is a gateway?"}])

const reply = await client.chat.completions.create({ model: "gpt-6.1-sol", messages: [{ role: "user", content: "What is a gateway?" }] });

Response

A Chat Completions response, or a stream of chunks with "stream": true. usage is always included; for streams the last chunk carries it. The x-sov-model header names the model that answered.

What Sovereign adjusts

Field	When	What happens
`model`	always	Set to the model chosen for the request.
effort	the chosen model reasons	Set to the effort the request needs, in the field the provider takes: `reasoning_effort` at OpenAI and Google, `reasoning.effort` through OpenRouter.
`max_tokens`	sent to OpenAI's own API	Sent as `max_completion_tokens`, the field OpenAI's newer models take.
`stream_options.include_usage`	streaming	Turned on.