{"slug": "gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now", "title": "GPT-4o and GPT-4.1 Retire January 14 — Migrate to Luna Now", "summary": "OpenAI will retire GPT-4o, GPT-4o mini, GPT-4.1, and GPT-4.1 mini from its API on January 14, 2027, with new fine-tuning jobs on those base models freezing on January 6, 2027, according to the company's deprecation notice. The replacement model, gpt-6-luna, costs $0.10 per million input tokens and $0.50 per million output tokens — 96% less than GPT-4o's $2.50/$10.00 — with a 1,050,000-token context window versus GPT-4o's 128,000 and a composite quality index 167% higher. Migration requires removing the temperature and top_p parameters, which Luna rejects, and switching function-calling requests with reasoning enabled from /v1/chat/completions to /v1/responses.", "body_md": "OpenAI filed the paperwork today: GPT-4o, GPT-4o mini, GPT-4.1, and GPT-4.1 mini are being pulled from the API on **January 14, 2027**. If you have fine-tuned models on any of these, the window is tighter — new training jobs freeze on **January 6**. The replacement is [gpt-6-luna](https://developers.openai.com/api/docs/deprecations), which is 96% cheaper and outperforms GPT-4o on quality benchmarks. Here is what you need to change and when.\n\n## Which Models Are Affected\n\nThese are the specific model IDs going dark on January 14:\n\n- `gpt-4o-2024-08-06`\n- `gpt-4o-2024-11-20`\n- `gpt-4o-mini-2024-07-18`\n- `gpt-4.1-2025-04-14`\n- `gpt-4.1-mini-2025-04-14`\n\nAny fine-tuned model built on these bases follows the same schedule. If you hardcoded a snapshot ID in your application, your code breaks on that date. If you used a generic alias like `gpt-4o`, OpenAI will auto-redirect you — but verify that behavior in your environment before trusting it.\n\n## Two Dates That Matter\n\nMost coverage will focus on January 14. The date that matters more if you own fine-tuned models is earlier: **January 6, 2027** is the last day to create new fine-tuning jobs on any of the deprecated base models. After that, you can still run inference on existing fine-tuned models until the 14th — but you cannot retrain or create new variants.\n\nWrite both dates down. Miss January 6 and you have no path to iterate your fine-tuned model before the lights go out.\n\n## GPT-6 Luna: The Replacement\n\nLuna is not just the default fallback — it is genuinely better across most dimensions. At $0.10 per million input tokens and $0.50 per million output tokens, it costs 96% less than GPT-4o ($2.50/$10.00). Its context window is 1,050,000 tokens, roughly eight times larger than GPT-4o’s 128,000. On the composite quality index, it scores 167% higher than GPT-4o. See the [full benchmark breakdown](https://www.activepieces.com/blog/gpt-6-luna-pricing-benchmarks-and-performance-in-2026) for a detailed comparison.\n\nThe one real tradeoff is streaming latency. Luna is a reasoning model — it thinks before it outputs, so per-token streaming is slower than GPT-4o. For interactive, low-latency applications, test carefully. For batch processing, agents, and anything where quality matters more than time-to-first-token, Luna is the obvious upgrade.\n\n## How to Migrate\n\nOpenAI’s [official migration guide](https://help.openai.com/en/articles/20001600-migrating-from-gpt-4x-to-gpt-6-luna) recommends four steps. Follow them in order and do not skip the baseline phase — you will need it to catch regressions.\n\n1. **Baseline.** Record a set of representative production requests and their current outputs. This is your ground truth.\n2. **Evaluate with Luna.** Run the same requests through`gpt-6-luna` with`reasoning_effort` set to`\"none\"` . This mode makes Luna behave closest to GPT-4x without engaging the reasoning engine.\n3. **Adjust.** Compare outputs. Tune prompts, adjust verbosity expectations, and check formatting assumptions. Most prompts migrate cleanly; some verbose outputs need trimming instructions added.\n4. **Gradual rollout.** Route 5–10% of production traffic to Luna, monitor quality metrics, latency, and error rates, then increase incrementally.\n\n## Breaking Changes to Catch Before You Ship\n\nTwo specific API parameters that will break your calls:\n\n**`temperature` and `top_p`:** Luna rejects both. If your code passes these parameters, your API call returns an error. Remove them from any request targeting Luna.\n\n**Function tools with reasoning enabled:** If you use function calling with any `reasoning_effort` setting other than `\"none\"`, you must switch from `/v1/chat/completions` to `/v1/responses`. Starting with `reasoning_effort: \"none\"` lets you stay on the completions endpoint during initial migration.\n\n```\n# Before (breaks January 14, 2027)\nresponse = client.chat.completions.create(\n    model=\"gpt-4o-2024-08-06\",\n    temperature=0.7,  # Remove this\n    top_p=0.9,        # Remove this too\n    messages=[{\"role\": \"user\", \"content\": prompt}]\n)\n\n# After (working Luna call)\nresponse = client.chat.completions.create(\n    model=\"gpt-6-luna\",\n    # reasoning_effort=\"none\" for closest GPT-4x parity\n    messages=[{\"role\": \"user\", \"content\": prompt}]\n)\n```\n\n## If You Have Fine-Tuned Models\n\nThis is the group that needs to move fastest. Before January 6:\n\n1. Export your training data from the OpenAI platform\n2. Start a new fine-tuning job targeting `gpt-6-luna`\n3. Evaluate the new fine-tuned Luna model against your benchmarks\n4. Update your endpoint references from `ft:gpt-4.1-...` to the new fine-tuned Luna model ID\n\nFine-tuning on Luna is available now. There is no technical reason to wait — and the cost of training is lower given Luna’s pricing. The only reason people will miss January 6 is procrastination.\n\n## Start Now\n\n91 days sounds like enough time until your Q4 roadmap fills up. The migration itself is straightforward — OpenAI did the work to make Luna the clear upgrade path. What takes time is evaluating output quality, catching the `temperature`/` top_p` parameter errors in your codebase, and retraining any fine-tuned models. Check the [GPT-4o vs Luna side-by-side comparison](https://www.orcarouter.ai/compare/gpt-4o-vs-gpt-6-luna) if you need to build a case internally for the switch.\n\nThe [official deprecation page](https://developers.openai.com/api/docs/deprecations) lists all affected models and dates. January 14 is a hard cutoff with no grace period.", "url": "https://wpnews.pro/news/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now", "canonical_source": "https://byteiota.com/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now/", "published_at": "2026-10-14 21:10:35+00:00", "updated_at": "2026-10-15 00:27:35.983541+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "developer-tools", "ai-infrastructure"], "entities": ["OpenAI", "GPT-4o", "GPT-4o mini", "GPT-4.1", "GPT-4.1 mini", "gpt-6-luna"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now", "markdown": "https://wpnews.pro/news/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now.md", "text": "https://wpnews.pro/news/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now.txt", "jsonld": "https://wpnews.pro/news/gpt-4o-and-gpt-4-1-retire-january-14-migrate-to-luna-now.jsonld"}}