Quickstart
Create a key, send a request, and see it in your dashboard.
1. Connect a provider and create a key
An admin connects your company's provider key once, on the Keys page: OpenRouter for every model, or a provider's own key, like OpenAI's or Anthropic's, for its models. Setting up a whole company? Follow Set up your company, with a screenshot of every step.
Then, on the same page, create a key for yourself (a person) or for a service (an app). Put it in a team if you use budgets. The key is shown once: copy it now.
export SOV_API_KEY="sov_live_…"
2. Send a request
Use any model name your tools already use. Sovereign starts from it and answers on the right model.
curl https://sov.solibra.ai/v1/chat/completions \
-H "Authorization: Bearer $SOV_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-sonnet-5-5",
"messages": [{"role": "user", "content": "Summarize our refund policy in two sentences."}]
}'
import os
from openai import OpenAI
client = OpenAI(base_url="https://sov.solibra.ai/v1", api_key=os.environ["SOV_API_KEY"])
reply = client.chat.completions.create(
model="claude-sonnet-5-5",
messages=[{"role": "user", "content": "Summarize our refund policy in two sentences."}],
)
print(reply.choices[0].message.content)
import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://sov.solibra.ai/v1", apiKey: process.env.SOV_API_KEY });
const reply = await client.chat.completions.create({
model: "claude-sonnet-5-5",
messages: [{ role: "user", content: "Summarize our refund policy in two sentences." }],
});
console.log(reply.choices[0].message.content);
3. See which model answered
Every response carries an x-sov-model header with the model that answered. The request also shows up in your
dashboard under Sessions within a few seconds, with its cost and what it saved.
curl -s -D - -o /dev/null https://sov.solibra.ai/v1/chat/completions \
-H "Authorization: Bearer $SOV_API_KEY" -H "Content-Type: application/json" \
-d '{"model": "claude-sonnet-5-5", "messages": [{"role": "user", "content": "hi"}]}' | grep -i x-sov-model
Next
- Set up Claude Code, Codex or your framework.
- Read how models are chosen and how to group requests into conversations.