WebSocket support for OpenAI Responses API live on AI Gateway
40% faster. That's what Vercel's AI Gateway just delivered for agentic workflows with WebSocket support on OpenAI's Responses API.
Why it matters
Vercel shipped WebSocket mode for the Responses API on AI Gateway, enabling persistent connections that reduce overhead for multi-turn agentic applications. This is a backend infrastructure optimization that directly impacts latency and cost for production AI agents at scale.
The key facts
6 to knowWebSocket support now live on Vercel AI Gateway for OpenAI Responses API
~40% faster end-to-end execution on WebSockets for agentic rollouts with 20+ tool calls
Persistent connection eliminates need to re-send full context on each turn
Compatible with store=false and Zero Data Retention modes
Follows OpenAI's upstream specification
GET /v1/responses route accepts raw response.create frames over WebSocket
Go to the source
Vercel Blogvercel.com
Publisher excerpt: 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 , instead of re-sending the full context over a fresh HTTP request every turn.AI Gatewayprevious_response_id up to ~40% faster end-to-end…