🔍 Read the full analysis: A Practical Guide To OpenAI’s GPT-4o, GPT-4.1, And GPT-5-nano API Shutdown on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
OpenAI’s published deprecation schedule lists GPT-5, GPT-5 mini and GPT-5-nano for API shutdown on Dec. 11, 2026, followed by GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini on Jan. 14, 2027. The replacement models can differ in price and behavior, so developers should inventory model IDs and test their applications before migrating.
OpenAI plans to retire GPT-5, GPT-5 mini and GPT-5-nano from its API on Dec. 11, 2026, then remove specified GPT-4o and GPT-4.1 models on Jan. 14, 2027, according to the company’s deprecation schedule reviewed Nov. 1. Developers using those model IDs will need to migrate or risk failed API calls; the listed replacements may also change costs and application behavior.
The Dec. 11 schedule covers gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-pro, o3 and o3-pro. OpenAI lists GPT-5.6 variants as replacements: gpt-5.6-sol for GPT-5 and o3, gpt-5.6-terra for GPT-5 mini, and gpt-5.6-luna for GPT-5-nano. The company also lists the pro reasoning mode for the recommended replacement of GPT-5-pro and o3-pro.
On Jan. 14, the listed retirements include GPT-4o snapshots, GPT-4o mini, GPT-4.1 and GPT-4.1 mini. OpenAI recommends gpt-6-luna for those models. The article source says older IDs, including gpt-4.1-nano, the original GPT-4o snapshot, GPT-4 Turbo, GPT-3.5 Turbo, o1 and o4-mini, were scheduled to stop working on Oct. 23. Developers should check the current deprecation page rather than assume an older integration still works.
The date list names specific model snapshots, while common short names such as gpt-4o and gpt-4.1 are marked deprecated in their model pages, according to the source. OpenAI says fine-tuned versions follow the retirement schedule of their base models. The source reports that the listed replacement models do not support fine-tuning, which may affect teams relying on custom-trained versions.
Infographic 01 · The shutdown calendar
What stops working, and when
OpenAI's API deprecations, as listed on 1 November 2026. GPT-4o and GPT-4.1 already left ChatGPT on 13 February 2026; this round is about apps built on the API.
- 23 Oct 2026Already offAlready shut down
gpt-3.5-turbo,gpt-4,gpt-4-turbo,gpt-4.1-nano,gpt-4o-2024-05-13,o1,o1-pro,o3-mini,o4-mini,gpt-image-1. Calls to these fail today. - 30 Nov 2026Your testsEvals platform, Agent Builder, reusable prompts
Not models, but tools many teams test with. Existing evals became read-only on 31 Oct. Export them first.
- 11 Dec 2026Notified 11 JunGPT-5 family and o3
gpt-5→gpt-5.6-sol·gpt-5-mini→gpt-5.6-terra·gpt-5-nano→gpt-5.6-luna·o3→gpt-5.6-sol·gpt-5-pro,o3-pro→gpt-5.6-sol(pro mode) - 14 Jan 2027Notified 14 OctGPT-4o, GPT-4o mini, GPT-4.1, GPT-4.1 mini
Snapshots
gpt-4o-2024-08-06,gpt-4o-2024-11-20,gpt-4o-mini-2024-07-18,gpt-4.1-2025-04-14,gpt-4.1-mini-2025-04-14→gpt-6-luna. Fine-tuned versions follow their base model. - Jan to Apr 2027Plan aheadAudio, speech, older GPT-5.x
6 Jan:
tts-1and others · 20 Jan: older audio and realtime models · 26 Feb:whisper-1, GPT-4o transcription · 1 Apr:gpt-5.1,gpt-5.4-nano,gpt-5.3-codex
Sources: OpenAI deprecations page (checked 1 Nov 2026) · OpenAI Help Center, model release notes. Dates as listed by OpenAI; shutdowns can be rescheduled.
Powered by Thorsten Meyer AI
Migration Risks for API Customers
A model shutdown can stop a production feature from responding if an application continues to send requests to a retired ID. The impact may extend beyond editing a configuration value: teams need to check prompts, output formats, latency, evaluation results and any fine-tuned models tied to the retiring base.
Price changes are not uniform. The source’s comparison of standard short-context rates, checked Nov. 1, says GPT-4o is listed at $2.50 per million input tokens and $10 per million output tokens, compared with $0.10 and $0.50 for gpt-6-luna. In contrast, GPT-5-nano’s listed replacement, gpt-5.6-luna, is priced at $0.20 per million input tokens and $1.20 per million output tokens, versus $0.05 and $0.40 for GPT-5-nano. These are stated prices, not a forecast of each customer’s bill; usage patterns, caching and prompt length affect totals.
The replacement models are described in the source as reasoning models. It warns that hidden reasoning tokens are billed as output, some sampling parameters must be removed, and a restrictive output-token limit could produce an empty answer that still incurs a charge. Those points make migration testing a reliability and budgeting task, not just a model-name change.
Popular right now
Vitamix Propel Series 750 Professional-Grade Blender, 4 Programs, Black
Ninja DualZone Smart XL Air Fryer, 10-Qt, 6-in-1, DZ950
Ninja NC501 CREAMi XL Deluxe 11-in-1 Ice Cream and Frozen Treat Maker with Two XL Containers
As an affiliate, we earn on qualifying purchases.
A Staged Retirement Calendar
The schedule also includes Nov. 30, 2026, when OpenAI plans to shut down its Evals platform, Agent Builder and reusable prompt objects. The source says existing evals became read-only on Oct. 31. Teams that use those tools to compare model behavior may need to preserve or replace their testing process before the shutdown.
Further retirements are listed for 2027, including text-to-speech models on Jan. 6, older audio and realtime models on Jan. 20, and Whisper and GPT-4o transcription models on Feb. 26. The source also lists gpt-5.1, gpt-5.4-nano and gpt-5.3-codex for April 1. These dates are based on the deprecation page as described by the source and should be rechecked as OpenAI updates its notices.
The source says OpenAI removed GPT-4o, GPT-4.1 and GPT-4.1 mini from ChatGPT on Feb. 13, 2026. That was a separate product change: the current schedule concerns API access, which supports third-party applications, internal tools and automated workflows. OpenAI’s deprecation page says affected customers are emailed and that generally available models receive at least six months’ notice unless safety or compliance concerns require a faster timeline. The source notes that the Jan. 14 notice was dated Oct. 14, but the entry does not explain how the notice-period policy applies.
““The replacements think before they answer.””
— Thorsten Meyer, ThorstenMeyerAI.com
Open Questions Before the Deadlines
The deprecation schedule provides dates and recommended replacements, but the source does not establish whether OpenAI will revise any dates or model recommendations. It also does not explain why the Jan. 14 retirements appear to have a shorter notice period than the general six-month policy; OpenAI’s entry does not state whether an exception applies.
Actual migration costs will depend on each application’s token use and configuration. The source provides standard short-context prices, not a customer-specific estimate, and those prices may change. It also says some customers may be able to buy dedicated capacity to keep access after shutdown, but does not give eligibility, availability or pricing details.
The source describes a Model Spec revision dated Oct. 29 that removes most detailed style guidance, including a default preference for Markdown, and instructs models to interpret instructions by their intent. It does not establish how a particular application will behave after model migration or under the revised guidance. Those effects need to be tested against each product’s requirements.
Inventory, Test and Confirm
Teams should first search application code, configuration files, automation platforms and test fixtures for every model ID, including dated snapshots and short names. Mark IDs already past their scheduled date for immediate checks, then prioritize the Nov. 30 tooling closure and Dec. 11 model retirements ahead of the January deadline.
Next, run representative requests against the proposed replacement and compare results with the current system. Test required formats, tool calls, safety behavior, latency and output limits; confirm that sampling parameters are supported. Measure token use and costs with realistic workloads, including any hidden reasoning tokens billed as output, and update monitoring so failed calls and unexpected empty responses are visible.
OpenAI’s schedule and pricing should be checked again before deployment. Customers with contractual commitments, questions about the notice period or interest in dedicated capacity can ask their OpenAI account team for details. The immediate next milestone is Nov. 30, followed by the Dec. 11 retirements; the source does not report a separate migration deadline or a guarantee that replacement behavior will match the retiring models.
Key Questions
When do GPT-5-nano and the other listed GPT-5 models leave the API?
OpenAI’s schedule, as reported in the source, lists Dec. 11, 2026 for GPT-5, GPT-5 mini, GPT-5-nano, GPT-5-pro, o3 and o3-pro.
When are GPT-4o and GPT-4.1 API models scheduled to shut down?
The specified GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are listed for Jan. 14, 2027. Check the live deprecation notice for the affected IDs and any updates.
Will the replacement models cost less?
It depends on the model. The source’s listed short-context prices show gpt-6-luna below GPT-4o’s standard per-token rates, while gpt-5.6-luna is more expensive per token than GPT-5-nano. A customer’s total depends on usage and current pricing.
What should developers test before switching?
Check output format, prompts, tool use, supported sampling parameters, token limits, response quality, latency and billing. Also verify whether a fine-tuned model is affected, since the source says fine-tuned versions follow their base model’s retirement schedule.
Source: ThorstenMeyerAI.com
