{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiepzs2btl4oe546yljnejgexedbqn3ihoeegyvlvpowtdt3uqnuhe",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlvvnhrhgns2"
  },
  "path": "/t/frustrated-with-new-codex-limits-for-the-pro-users/1380784#post_12",
  "publishedAt": "2026-05-15T17:40:17.000Z",
  "site": "https://community.openai.com",
  "textContent": "Serdjio:\n\n> At minimum, the UI should explain the limits honestly, and there should be some fallback/basic model for audits, tests, or smaller tasks after the main quota is exhausted.\n\nThis honestly hadn’t occurred to me until you pointed out the problem, but anyone dealing with a standard IT sysadmin will recognise this: Q: “why didn’t you tell me that in the first place??” A: “because you didn’t ask?”\n\nNext time I use codex-cli I will ask it the question, “if i use 5.5 for this task, will you run out of ‘tokens’ before we are even done?”\n\nLet’s just pray the answer isn’t:\n\nA: “i’m sorry, you’ve reached your token limit by asking that question”\n\nWhich… yeah. There should be clear costs with estimates up-front.",
  "title": "Frustrated with new Codex Limits for the Pro Users"
}