Menu
Sign inGet started

Build with the API

Quickstart: your first API request

This page takes you from nothing to a working request. You need an account with credit and a way to call it: curl, or the Python or Node client you already use.

1. Create a key#

Sign up if you have not, add credit under Billing, then open API keys and choose Create. Give the key a name, such as build-server, and copy it when it is shown: it is shown once. Keep it out of source code and put it in your environment.

bash
export SVEN_API_KEY="paste-your-key-here"

Only an admin of an organisation can create keys; in a personal account you can. See Authentication and API keys.

2. Install a client#

Skip this for curl. The clients you already have for chat work as they are: this API is the one they talk to once you change their base URL and key.

pip install openai

3. Choose a model#

List the models your key can call and pick a chat model's id. Prices are on the pricing page.

curl https://sven.swedishembedded.com/v1/models \
  -H "Authorization: Bearer $SVEN_API_KEY"

In the examples below, replace MODEL_ID with the id you chose.

4. Send your first request#

curl https://sven.swedishembedded.com/v1/chat/completions \
  -H "Authorization: Bearer $SVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "MODEL_ID", "messages": [{"role": "user", "content": "Hello"}]}'

The answer is the usual chat completion object. Your charge appears in Usage and logs as soon as the call finishes.

5. Stream the answer#

Add stream and the reply arrives as it is written.

curl -N https://sven.swedishembedded.com/v1/chat/completions \
  -H "Authorization: Bearer $SVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MODEL_ID",
    "messages": [{"role": "user", "content": "Write a haiku about the sea"}],
    "stream": true
  }'

More in Streaming.

Where to go next#