{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidekn3cytaz2v3hoviqjp5b7fkhbsak75udbrnd6v5zyakrodzlfa",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mkgykonxbjc2"
  },
  "path": "/t/developing-sprite-sheets-with-gpt-image-2/1379831#post_5",
  "publishedAt": "2026-04-27T00:11:44.000Z",
  "site": "https://community.openai.com",
  "textContent": "Original approach with dummy model in sprites and then “character skin” applied by model seems easier for me if production scale is needed.\n\nLooks similar to my concept of \" brand voice skin\" where model converts brand text to \" neutral \" and then another one gets trained on reversed samples to convert\" neutral\" in \" branded\".\n\nMy gut says here the “neutral” is the dummy and the approach might actually work way better than one may think. Just some fine-tuning will definitely be necessary if truly going to scale volumes.",
  "title": "Developing sprite sheets with gpt-image-2"
}