{"id":"2049595890395152728","url":"https://x.com/OpenAIDevs/status/2049595890395152728","text":"⚙️ We made agent loops faster with WebSockets in the Responses API\n\nAs Codex got faster, the bottleneck moved from inference to inefficient API calls\n\nWebSockets keep response state warm across tool calls, helping workflows run up to 40% faster end to end\n\nhttps://openai.com/index/speeding-up-agentic-workflows-with-websockets","author":{"name":"OpenAI Developers","username":"OpenAIDevs","avatarUrl":"https://pbs.twimg.com/profile_images/2022002720971096064/l3Kyt4qt_200x200.jpg"},"createdAt":"Wed Apr 29 21:05:14 +0000 2026","engagement":{"replies":33,"retweets":74,"likes":1253,"views":245157},"externalLink":{"url":"https://openai.com/index/speeding-up-agentic-workflows-with-websockets","displayUrl":"openai.com","title":"Speeding up agentic workflows with WebSockets in the Responses API","description":"A deep dive into the Codex agent loop, showing how WebSockets and connection-scoped caching reduced API overhead and improved model latency.","thumbnailUrl":"https://images.ctfassets.net/kftzwdyauwt9/4OePmYEOLzZNcxVLLd9pMH/cc16f505aa589ef72ef3c1c265705727/SEO.png?w=1600&h=900&fit=fill"},"adhxContext":{"savedByCount":1,"publicTags":[],"previewUrl":"https://adhx.com/OpenAIDevs/status/2049595890395152728"}}