{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreih2vzoqaipxq46avrnzqitrkwggwhlkvmelf3y34s33ttxjadowem",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mldf4jpmjw72"
  },
  "path": "/t/is-this-a-world-record-the-gpt-5-5-ran-for-19-hours/1380472#post_9",
  "publishedAt": "2026-05-08T07:14:07.000Z",
  "site": "https://community.openai.com",
  "textContent": "Thanks for sharing this, Sam. That actually makes sense.\n\nI left the agent running on its own and went to sleep. Judging by the finish time, it worked for almost a full 24 hours.\n\nWhen I asked it to check how much of the original task was actually completed, it estimated around 65–70%. So I guess it needs another 24 hours.\n\nWhat you wrote about `/goal` explains the behavior pretty well. It feels less like a single huge model response and more like a goal-driven runtime that can keep working over a long session.\n\nFor my project, though, I’m using GPT-5.5 on extra high reasoning instead of medium. The codebase is around 1.4 million lines now, so I don’t really want to risk it. In these long autonomous runs, one wrong decision can hurt.\n\nAbout 30% of the project was built with GPT-5.3, 40% with GPT-5.4, and the last 30% with GPT-5.5. At this scale, medium reasoning feels like a risk I don’t need to take.",
  "title": "Is this a world record? The GPT 5.5 ran for 19 hours"
}