Intel / article

WebSocket support for OpenAI Responses API live on AI Gateway

Source: vercel.com
Author: Kevin Dawkins
Date: Published Jul 27, 2026

Why it matters

If you run tool-call-heavy agents through OpenAI's Responses API, this removes per-turn HTTP setup cost by holding a persistent WebSocket and letting you send only new input plus a previous_response_id — with store=false and Zero Data Retention still supported, so latency gains don't force you into data retention tradeoffs.

Key quotes

“AI Gateway now supports WebSocket mode for the OpenAI Responses API.”

“You can keep a persistent connection open and continue each turn by sending only new input items plus previous_response_id , instead of re-sending the full context over a fresh HTTP request every turn.”

“OpenAI reports up to ~40% faster end-to-end execution on WebSockets for agentic rollouts with 20 or more tool calls.”

Read the source vercel.com