{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreif3htybp6r4lfwryga3gt6zgmhfn7uzwpw3k3spy53qgdhpfh4nbi",
"uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mk4p4g7qreh2"
},
"path": "/t/deprecation-of-fine-tuned-models-but-still-cant-access-newer-ones/1379550#post_4",
"publishedAt": "2026-04-22T23:01:39.000Z",
"site": "https://community.openai.com",
"tags": [
"developers.openai.com/api/docs/deprecations",
"supervised fine-tuning",
"direct preference optimization",
"reinforcement fine-tuning"
],
"textContent": "> Hello,\n>\n> _You’re not misreading the shutdown notice: fine-tuned versions of_ _gpt-4.1-nano-2025-04-14_ _are scheduled to be removed on October 23, 2026, and the deprecations page lists_ _gpt-5-nano_ _as the recommended substitute._\n>\n> The “substitute model” column is a migration target for using the API after shutdown; it does not currently mean that _gpt-5-nano_ _is available for fine-tuning. Fine-tuning itself is not being removed, but GPT-5-family models are not currently listed in the public fine-tuning docs as supported fine-tuning base models._\n>\n> If you need a fine-tuned replacement today, please use one of the currently supported fine-tuning models documented in the fine-tuning guides. If your goal is specifically to migrate from a fine-tuned _gpt-4.1-nano_ _model, we recommend evaluating base_ _gpt-5-nano_ _against your existing evals/prompts, since that is the documented substitute for that deprecated model family._\n\nReference:\n\n * Deprecation page: developers.openai.com/api/docs/deprecations\n * SFT docs: supervised fine-tuning\n * DPO docs: direct preference optimization\n * RFT docs: reinforcement fine-tuning\n\n",
"title": "Deprecation of fine tuned models, but still can't access newer ones?"
}