{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreib4lwfmzsw422skqng7zgupxk7nafxr6w3w6yktyilya3dgi7lziy",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3ml53lismng62"
  },
  "path": "/t/please-don-t-retire-gpt-5-1-thinking-gpt-5-2-feels-worse/1375332?page=10#post_197",
  "publishedAt": "2026-05-05T20:23:46.000Z",
  "site": "https://community.openai.com",
  "textContent": "Do you use the web interface or the mobile app? Just to be sured, check which model actually generated your response. I was a little freaked out just now—I was messaging in a few chats asking for a 5.5 Thinking score, and in the cases where I thought the response was less accurate, I see that a 5.4 was generated. That’s really strange.\n\nOkay, this is just ridiculous—now it keeps showing up in my recent posts that it was 5.4, even though I’m sure I set it to 5.5. The UI is a total mess.",
  "title": "Please don’t retire GPT-5.1 Thinking – GPT-5.2 feels worse"
}