Skip to main content

Streaming Chat

Streaming Chat enables your agent to receive model responses incrementally — as they are generated — instead of waiting for the full response.

This is ideal for:

  • Live agent interactions
  • Interactive UIs with typing indicators
  • Reducing latency for long responses

Current Status​

Streaming is not yet supported in Telex AI.

This page describes the future planned behavior so agents can prepare for upcoming updates.

🔗 Endpoint​

POST /telexai/chat

Required Headers​

  • X-AGENT-API-KEY: Your API key

  • Model can be specified via:

    • X-Model header
    • model query param
    • model in request body

Planned Example​

{
  "messages": [
    {
      "role": "user",
      "content": "Write a poem about the stars."
    }
  ],
  "stream": true
}

Planned Response: Server-Sent Events (SSE) stream of message chunks.

[data stream begins]

{"role": "assistant", "content": "The stars whisper softly..."}
{"role": "assistant", "content": "... in the midnight sky."}
...
[data stream ends]

Usage Notes​

  • Set "stream": true in request body to enable
  • Future response will use SSE for compatibility with browsers and real-time apps
  • Until officially supported, fallback to standard synchronous responses

Next​