{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiftqbol4hgn6yw4v7p3akl2dl77mgyaaaegfi774goz3e2lvtdlgy",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mg6vcw2ex6n2"
  },
  "path": "/t/error-when-attempting-to-access-gpt-image-1-5-in-images-edits/1375534#post_15",
  "publishedAt": "2026-03-03T21:51:15.000Z",
  "site": "https://community.openai.com",
  "textContent": "What am I missing?\n\n_j:\n\n> The format in which generated images with `dall-e-2` and `dall-e-3` are returned. … This parameter isn’t supported for `gpt-image-1` which will always return base64-encoded images.\n\nWhat I understand is that there is one endpoint each for image creation and image edits, but the supported models require different parameters. That is confusing. Especially because all parameters suddenly appear optional and are listed both in the request reference and in the created object. One can find all the information in the descriptions but it’s easy to miss something here.\n\nIn addition, the documentation for the DALL·E models appears to be temporarily missing, and the response_format parameter is also missing from the Image Edits API reference.",
  "title": "Error when attempting to access gpt-image-1.5 in images/edits"
}