gpt.buzz

news

Speeding up agentic workflows with WebSockets in the Responses API

April 22, 2026

A deep dive into the Codex agent loop, showing how WebSockets and connection-scoped caching reduced API overhead and improved model latency.

A deep dive into the Codex agent loop, showing how WebSockets and connection-scoped caching reduced API overhead and improved model latency.

Source: openai.com

← All news