How do I keep a conversation going in Python?

The Model API doesn't remember earlier requests. Keep a messages list, append the assistant's reply after each call, and send the full list next time.

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.gateway.clatterlabs.xyz/v1",
    api_key=os.environ["CLATTER_API_KEY"],
    default_headers={"X-Clatter-Account": os.environ["CLATTER_ACCOUNT_ID"]},
)

messages = [{"role": "system", "content": "You are a friendly tutor."}]

def ask(text):
    messages.append({"role": "user", "content": text})
    resp = client.chat.completions.create(model="supercosmos-flash", messages=messages)
    reply = resp.choices[0].message
    messages.append({"role": "assistant", "content": reply.content})
    return reply.content

print(ask("What is a prime number?"))
print(ask("Is 51 one?"))

History counts as input tokens every turn, so long chats cost more per turn. A request can hold up to 800 messages and 3 MB; trim or summarise old turns well before that.

The client setup above (base URL, key and the X-Clatter-Account header) is the same in every Python example; see Set up the Python SDK. Full parameter list: Create a chat completion.

Good to know#

• Retry only 429 and 5xx errors, with backoff; honour Retry-After when it's sent. Which errors to retry
• Every response has an X-Request-Id; log it so a request can be found later. Why log it
• Each key has its own limit: 10, 30 (default), 60 or 120 requests per minute. Rate limits

Still need help?#

• General support: support@clatterlabs.com or team@clatterlabs.xyz
• Infrastructure questions, or permanent deletion of your data (your whole account or one service): infrastructure@clatterlabs.xyz

We usually reply within the first hour, and always within 24 hours. A full support team is with you.