API Reference¶

Illinois Chat exposes a public REST API so you can integrate your assistants into your own applications. All requests and responses are JSON over HTTPS.

Base URL¶

https://chat.illinois.edu

For self-hosted deployments, substitute your own host. The chat endpoint is served by the Next.js frontend; retrieval, ingest, and export endpoints are served by the Flask backend (on self-hosted stacks these run on different ports — frontend :3000, backend :8000).

Endpoint categories¶

Category	What it does	Docs
Chat	RAG-grounded, multi-turn conversations with streaming and image support	Chat
Retrieval	Fetch relevant document contexts without invoking an LLM	Retrieval
Ingest	Add files, URLs, or Canvas courses to a project programmatically	Ingest
Export	Bulk-export documents and conversation history	Export

Authentication¶

Requests are authenticated with a project API key passed in the JSON body (not a header), plus your own LLM provider key when a commercial model is used. See Authentication.

Streaming¶

Endpoints that support streaming ("stream": true) return newline-delimited chunks as the model generates them.

Rate limits¶

No rate limits are currently published for the hosted instance. Be considerate; heavy programmatic workloads are better suited to a self-hosted deployment.

Quick example¶

curl -X POST https://chat.illinois.edu/api/chat-api/chat \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Summarize the key topics in these documents."}],
    "course_name": "your-project-name",
    "api_key": "uc_YOUR_API_KEY",
    "openai_key": "sk-YOUR_PROVIDER_KEY",
    "temperature": 0.1,
    "stream": false
  }'