Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.lyceum.technology/llms.txt

Use this file to discover all available pages before exploring further.

For long-running inference where you want progress as it happens — rather than waiting for the full response — Lyceum Cloud exposes a Server-Sent Events (SSE) endpoint per inference request.

​
Endpoint

GET /api/v2/external/inference/streaming/stream/{inference_id}
The stream emits a queued event immediately, then either finished or failed when the inference resolves.

​
When to use this

  • You want to display progress to a user during a long inference
  • You’re building a chat UI where partial responses should appear as they’re generated
  • You’re integrating with a frontend that already speaks SSE (most do)
For a one-shot blocking call where you only need the final response, the regular chat completions endpoint is simpler.

​
Consuming the stream

import requests

inference_id = "..."
url = f"https://api.lyceum.technology/api/v2/external/inference/streaming/stream/{inference_id}"

with requests.get(
    url,
    headers={"Authorization": f"Bearer {api_key}"},
    stream=True,
) as resp:
    for line in resp.iter_lines():
        if line:
            print(line.decode())
Any standard SSE client works the same way — EventSource in browsers, sseclient-py, etc.