🔍 Read the full analysis: From GPT-4o To A New API Model: How To Plan Your OpenAI Migration on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI’s published API deprecation schedule lists shutdowns on Dec. 11, 2026, and Jan. 14, 2027, affecting GPT-5, GPT-4o and other models; several IDs were already scheduled to stop working on Oct. 23. A Nov. 1 article by Thorsten Meyer says the listed replacements differ in price and behavior, so developers should test model changes and verify current details with OpenAI.
OpenAI’s API deprecation schedule lists model shutdowns for Dec. 11, 2026, and Jan. 14, 2027, affecting GPT-5, GPT-4o, GPT-4.1 and related models. The dates matter to companies using those model IDs in software: requests may stop working after retirement, and the replacements listed by OpenAI can differ in price, response behavior and API settings.
Thorsten Meyer reported on Nov. 1 that the schedule’s Dec. 11 retirements include GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro. The listed replacements are in the GPT-5.6 family: GPT-5 and o3 map to GPT-5.6 Sol; GPT-5 mini to GPT-5.6 Terra; GPT-5 nano to GPT-5.6 Luna. GPT-5 pro and o3-pro are listed for GPT-5.6 Sol with its pro reasoning mode.
On Jan. 14, 2027, the schedule lists dated snapshots of GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini for retirement, with GPT-6 Luna named as the recommended replacement. Meyer says the model pages for the corresponding short names already showed “Deprecated” when he checked. He also reports that a group of older IDs, including GPT-4 Turbo, GPT-3.5 Turbo, o1 and o4-mini, had a listed shutdown date of Oct. 23, 2026.
The article’s pricing comparison, based on OpenAI’s standard short-context prices checked Nov. 1, shows that a migration can change costs in either direction. Meyer reports GPT-6 Luna at $0.10 per million input tokens and $0.50 per million output tokens, compared with GPT-4o at $2.50 and $10. For GPT-5 nano users moving to GPT-5.6 Luna, he lists $0.20 input and $1.20 output, versus $0.05 and $0.40 for GPT-5 nano. These are reported prices at a specific date, not a guarantee of future rates.
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 Changes Cost and Behavior
Changing a model ID is not necessarily a drop-in update. Meyer describes the listed successors as reasoning models, which can use hidden reasoning tokens billed as output. He warns that some sampling parameters may need to be removed and that a tight output-token limit could result in an empty answer while still incurring a charge. Developers should validate actual usage and billing rather than infer them from a successful API call.
There is also a product consistency issue. The article says OpenAI’s Oct. 29 Model Spec revision removed much of its detailed style guidance, including a default preference for Markdown formatting, and tells models to interpret instructions by intent. If an application depends on a particular structure or tone, its prompt and tests should specify that requirement directly. Meyer notes that OpenAI says its models “do not yet fully reflect” the Spec, so the document should not be treated as a guarantee of identical behavior across models.
For organizations, the practical stakes include service continuity, user experience and budget. A chatbot that returns malformed data, changes tone or exceeds expected token costs can cause problems even when its replacement model is available. Migration planning needs both engineering checks and review of the outputs users actually see.
Popular right now
De'Longhi 15-Bar Pump Espresso Machine with Milk Frother & Temperature Control
De'Longhi Rivelia Automatic Espresso Machine with Milk Frother
Dyson Ball Animal 3 Upright Vacuum
As an affiliate, we earn on qualifying purchases.
A Staged API Retirement Schedule
The dates come from OpenAI’s deprecations page, which Meyer says he checked on Nov. 1, 2026. He reports that OpenAI says it emails affected customers and documents deprecations on that page. The article distinguishes API retirements from a prior ChatGPT change: it says GPT-4o, GPT-4.1 and GPT-4.1 mini were removed from ChatGPT on Feb. 13, 2026. The current schedule concerns API integrations, including applications and bots built by customers.
Meyer also flags a separate Nov. 30, 2026, shutdown affecting OpenAI’s Evals platform, Agent Builder and reusable prompt objects. Existing evals became read-only on Oct. 31, according to his account. That timing could complicate migration work for teams using those tools to test or manage prompts, so they may need to plan for changes to their evaluation workflow as well as model access.
The article says OpenAI’s notice page promises at least six months’ notice for generally available models, except where safety or compliance concerns call for a faster timeline. It notes that the Jan. 14 retirement notice was dated Oct. 14, leaving three months, but says the entry does not explain which rule applies. Meyer also reports that some customers may be able to purchase dedicated capacity for continued access after a shutdown.
Open Questions in the Schedule
The schedule’s model dates and suggested replacements are reported from OpenAI’s published information, but the reason for the three-month notice ahead of the Jan. 14 retirements is not explained in the material. It is also unclear which affected customers qualify for dedicated capacity after a model is retired, what it would cost, or how long access could continue.
The prices in Meyer’s comparison were checked on Nov. 1 and may change. The source also gives only part of the pricing and migration comparison, so teams should consult current pricing and model documentation before estimating their own costs. The article’s warnings about changed outputs and token limits are migration risks to test, not evidence that every application will experience those problems.
Audit, Test and Confirm Dates
Teams should inventory model IDs across source code, configuration files, test fixtures and no-code automations, then check each against OpenAI’s current deprecations page. The listed Oct. 23 retirements should be treated as immediate checks for any system still using those IDs. For upcoming dates, developers can migrate in a test environment and compare output quality, formatting, latency, token use and cost with representative requests.
Before switching production traffic, teams should review prompts and API parameters for assumptions that may not carry over, set output limits with the reasoning-model behavior in mind, and verify that automated evaluations still work. Customers with contracts or committed spend can ask their OpenAI account team about the notice period and possible dedicated capacity. OpenAI’s schedule and pricing should be rechecked before each migration, since the article reflects information available on Nov. 1, 2026.
Key Questions
When are GPT-4o API models scheduled to shut down?
The schedule reported by Thorsten Meyer lists dated GPT-4o snapshots for retirement on Jan. 14, 2027, with GPT-6 Luna named as the recommended replacement. Check OpenAI’s current deprecations page for the latest status of specific model IDs.
Which models are listed for retirement on Dec. 11, 2026?
The reported list includes GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro. The listed replacements are in the GPT-5.6 family, with mappings that vary by retiring model.
Will the replacement models cost less?
Not in every case. Meyer’s Nov. 1 comparison reports lower listed prices for some moves, including GPT-4o to GPT-6 Luna, but higher input and output prices for GPT-5 nano moving to GPT-5.6 Luna. Prices can change, so compare current rates using your application’s actual token use.
What should developers test before migrating?
Test representative prompts for output format, answer quality, token usage, costs and API parameter compatibility. The article says reasoning-model replacements may bill hidden reasoning tokens as output and that tight output limits can produce empty answers.
Are the API retirements the same as ChatGPT model removals?
No. The reported dates concern models used through the OpenAI API. Meyer separately says GPT-4o, GPT-4.1 and GPT-4.1 mini were removed from ChatGPT on Feb. 13, 2026.
Source: ThorstenMeyerAI.com
