🔍 Read the full analysis: A Timeline For OpenAI’s GPT-4o, GPT-4.1, And GPT-5-nano API Changes 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 API schedule lists model shutdowns on Dec. 11, 2026, and Jan. 14, 2027, with other model IDs already shut down on Oct. 23. The listed replacements may change costs and response behavior, but the source does not provide a full migration or testing plan, and some schedule details need clarification.
OpenAI’s API deprecation schedule sets Dec. 11, 2026, for GPT-5, GPT-5 mini, GPT-5-nano, o3 and related models to leave the API, followed by GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini on Jan. 14, 2027. The dates matter to developers whose applications call those model IDs: requests using a shut-down model may fail, while moving to a listed replacement can change price and output behavior.
ThorstenMeyerAI.com reported on Nov. 1 that its dates and model list came from OpenAI’s deprecations page. It said a group of models, including gpt-4.1-nano, gpt-4-turbo, gpt-3.5-turbo, o1 and o4-mini, was already shut down on Oct. 23. The article says calls using those IDs may now fail. Its schedule also lists the original GPT-4o snapshot, gpt-4o-2024-05-13, among the models removed that day.
For Dec. 11, the listed migrations include GPT-5 to gpt-5.6-sol, GPT-5 mini to gpt-5.6-terra, and GPT-5-nano to gpt-5.6-luna. The article says o3 is also assigned to gpt-5.6-sol, while GPT-5-pro and o3-pro are assigned to that model with its pro reasoning mode. On Jan. 14, the schedule names dated GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots, with gpt-6-luna listed as the replacement for each.
The report says the replacement options are reasoning models, which can affect API use as well as billing. It describes hidden reasoning tokens as billed output and warns that some sampling parameters must be removed. It also says a tight output-token limit can result in an empty answer that is still billed. These are practical migration concerns reported by the source; developers should check the current model and pricing documentation for their own use case.
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 Users
The schedule affects software that calls models through the OpenAI API, including products, internal tools and automated workflows. A shutdown is not just a model-name change: an application that still sends requests to an unavailable ID may stop producing responses. Replacing the ID without testing can also change a system’s output format, latency, token use or cost.
Prices in the report show why teams should compare their actual workloads before switching. For short-context requests, it lists GPT-4o at $2.50 per million input tokens and $10 per million output tokens, compared with gpt-6-luna at $0.10 input and $0.50 output. But it gives a different direction for GPT-5-nano: its listed replacement, gpt-5.6-luna, is priced at $0.20 input and $1.20 output, against $0.05 and $0.40 for GPT-5-nano. Those figures are the source’s prices checked Nov. 1, 2026, not a forecast of a particular customer’s bill.
Behavior changes may be as consequential as price changes. The report connects the replacements’ reasoning behavior to output-token billing and parameter compatibility. Applications that rely on a fixed response format or use sampling settings may require edits and regression tests. The source also reports a change to OpenAI’s public Model Spec, but that document is separate from the API shutdown schedule and does not by itself establish exactly how any specific model will respond.
Popular right now
Shark Rotator Lift-Away Upright Vacuum with HEPA Filter (NV501)
De'Longhi Rivelia Automatic Espresso Machine with Milk Frother
De'Longhi 15-Bar Pump Espresso Machine with Milk Frother & Temperature Control
As an affiliate, we earn on qualifying purchases.
How the Shutdown Calendar Developed
According to the report, OpenAI notified customers about the Dec. 11 models on June 11 and the January models on Oct. 14. The article says OpenAI’s deprecation page promises at least six months’ notice for generally available models unless safety or compliance concerns require a faster timeline. It notes the three-month period between the October notice and the January shutdown, but says the notice does not explain which rule applies. That discrepancy is a point for customers to raise with OpenAI, not proof of a policy violation.
The schedule extends beyond the headline models. The report lists a Nov. 30, 2026, shutdown for OpenAI’s Evals platform, Agent Builder and reusable prompt objects at v1/prompts; it says existing evals became read-only on Oct. 31. It also lists later 2027 dates for text-to-speech, audio, realtime and transcription models, as well as gpt-5.1, gpt-5.4-nano and gpt-5.3-codex on April 1. These dates are included in the source’s account of the deprecation page and should be checked against OpenAI’s live notices.
The report distinguishes the API changes from ChatGPT changes. It says OpenAI removed GPT-4o, GPT-4.1 and GPT-4.1 mini from ChatGPT on Feb. 13, 2026. The deadlines in this story concern API model access, meaning applications built on the API rather than ChatGPT alone.
Open Questions on Dates and Behavior
The source reports the dates and replacements from OpenAI’s deprecations page, but does not reproduce the full official notices or clarify every status. In particular, it says the January schedule covers dated snapshots while the shorter model names are marked “Deprecated,” and advises treating those names as part of the January deadline. Developers should confirm the exact IDs used in their projects against OpenAI’s current documentation rather than assume every alias behaves identically.
The reason for the three-month notice period before the January shutdown is not given in the report. Nor does the material establish whether OpenAI will grant exceptions, how dedicated-capacity access would work for a given customer, or whether every affected account has received a notice. OpenAI is said to allow some customers to seek dedicated capacity after a shutdown, but eligibility and terms are not specified.
The report describes price differences and possible reasoning-model behavior, but does not supply a complete feature-by-feature comparison, benchmark results, or a general migration procedure. The Model Spec changes also do not establish that every API response will change in a predictable way. Actual costs and outputs will depend on the model, request length, settings and application, so those details remain to be tested.
Deadlines and Migration Checks
The next listed platform milestone is Nov. 30, 2026, when the report says Evals, Agent Builder and reusable prompt objects will shut down. The model deadline after that is Dec. 11, followed by the GPT-4o and GPT-4.1 family deadline on Jan. 14, 2027. OpenAI’s deprecations page remains the stated source for checking whether those dates or model mappings have changed.
Before a deadline, API users can inventory model IDs across code, configuration files, tests and automation tools, then compare each ID with the official schedule. A migration test should verify that the replacement accepts the application’s parameters, returns the required format and handles the same representative prompts. Teams can also compare token use and cost under their own workloads, including cases with long inputs or strict output limits. These steps are practical implications of the reported changes, not a test plan published by OpenAI.
Customers with contracts, committed spend or a need for continued access can ask their OpenAI account team about the notice period and the reported dedicated-capacity option. The report does not say what answers OpenAI will provide or whether any deadline will be adjusted. Further model and API changes are listed for 2027, so teams should keep checking the official schedule after completing the immediate migration work.
Key Questions
When are GPT-5-nano and GPT-5 mini leaving the API?
The schedule reported by ThorstenMeyerAI.com lists Dec. 11, 2026. It names gpt-5.6-luna as the GPT-5-nano replacement and gpt-5.6-terra for GPT-5 mini.
When do GPT-4o and GPT-4.1 API models shut down?
The report lists Jan. 14, 2027, for specified GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots. It says gpt-6-luna is the listed replacement. Check OpenAI’s current deprecations page for the exact model IDs and status.
Have any of the listed models already stopped working?
According to the report, OpenAI shut down several IDs on Oct. 23, 2026, including gpt-4.1-nano, gpt-4-turbo, gpt-3.5-turbo, o1 and o4-mini. Requests using those IDs may fail; confirm the current status with OpenAI’s documentation.
Will switching models change API costs?
It can. The source’s Nov. 1 price comparison says some listed replacements cost less per token than the models they replace, while gpt-5.6-luna is listed at higher input and output prices than GPT-5-nano. Actual spending depends on token use and the current prices, which may change.
What should API developers check before migrating?
Identify every affected model ID, check its shutdown date and listed replacement, then test application settings, output format, token limits and cost using representative requests. The report specifically warns that reasoning tokens are billed as output and that some sampling parameters may need to be removed.
Source: ThorstenMeyerAI.com
