{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiabhpmrbvfisg4qx7itytfnuw76oq2ntmbwbeo524hg3pgkkt3qv4",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mfueuaeflxc2"
  },
  "path": "/t/fine-tuned-4o-errors-using-the-responses-api-with-image-input/1375316#post_1",
  "publishedAt": "2026-02-27T17:57:03.000Z",
  "site": "https://community.openai.com",
  "textContent": "For the past 1 day, we’ve been getting incorrect responses from the responses API. The response keeps spamming an empty array:\n\n{\n\n\n           \"id\": \"<id>\",\n\n\n\n           \"type\": \"message\",\n\n\n\n           \"status\": \"completed\",\n\n\n\n           \"content\": [\n\n\n\n             {\n\n\n\n               \"type\": \"output_text\",\n\n\n\n               \"annotations\": [],\n\n\n\n               \"logprobs\": [],\n\n\n\n               \"text\": \"\"\n\n\n\n             }\n\n\n\n           ],\n\n\n\n           \"role\": \"assistant\"\n\n\n\n         },\n\n\nWhile completely ignoring the specified json schema. The response also indicates no errors, but seems we’re getting billed for the 100k+ tokens it ends up using :)))\n\nAnyone else experiencing this issue? I’m using a fine-tuned 4o model. Not every request has this death spiral though; some do go through correctly.\n\nEdit: I am using image input",
  "title": "Fine-tuned 4o errors using the responses API (with image input)"
}