🔍 Read the full analysis: Will Your OpenAI Integration Break When These Models Switch Off? on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
OpenAI’s API deprecation schedule lists model shutdowns on Dec. 11, 2026, and Jan. 14, 2027, after several older model IDs stopped working on Oct. 23. The source article says replacements may change costs and response behavior, but its supplied pricing details are incomplete and should be checked against OpenAI’s current documentation.
OpenAI is scheduled to retire several models from its API on Dec. 11, 2026, followed by GPT-4o and GPT-4.1 model snapshots on Jan. 14, 2027, according to a Nov. 1 report citing the company’s deprecations page. The changes affect developers and businesses whose apps, workflows or bots call those model IDs; older listed models have already been shut down, the report says.
The schedule described in the report identifies Dec. 11 for GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro. It says GPT-5 is mapped to GPT-5.6-sol, GPT-5 mini to GPT-5.6-terra and GPT-5 nano to GPT-5.6-luna. The report lists GPT-5.6-sol, with pro reasoning mode, as the replacement for GPT-5 pro and o3-pro; o3 is also mapped to GPT-5.6-sol.
On Jan. 14, 2027, the listed GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled to leave the API. The report names GPT-6-luna as the recommended replacement for those models. It says several other model IDs, including GPT-4-turbo, GPT-3.5-turbo, o1 and o4-mini, were already shut down on Oct. 23. Developers still using an affected ID may find that calls no longer work, according to the article.
The report also flags changes beyond model availability. It describes the replacements as reasoning models, which may handle parameters and output limits differently. It says hidden reasoning tokens are billed as output and warns that strict output limits can produce empty answers that are still billed. The supplied source does not provide the complete technical or pricing documentation for each replacement, so teams should verify details with OpenAI before changing production systems.
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
How API Changes Affect Deployed Apps
A retired model ID can interrupt a feature anywhere it is embedded: a server-side configuration, test fixture, third-party automation or no-code workflow. The source article notes that some integrations may be old and not actively maintained, making a deadline easy to miss. Because the API powers products built by other companies, the impact can reach their customers even if those customers never interact with OpenAI directly.
Migration is not just a matter of changing a model name. The report says replacement models may differ in price, output behavior and supported parameters. It gives examples in which listed replacements cost less for some GPT-4 family calls but more for GPT-5 nano and GPT-5 mini. The source excerpt cuts off partway through its price comparison, however, so it does not support a complete cost assessment. Teams need to compare their own token use, prompt lengths and workloads against current official rates.
The practical risk is a combination of service disruption and changed answers. Even if requests continue to succeed, a new model may format responses differently or behave differently under existing instructions. That can affect structured outputs, customer support replies and automated steps that rely on predictable text.
Products Worth Considering
Amazon Kindle Scribe (32GB) with Premium Pen, Tungsten
LG UltraGear 32GX850A-B 32-inch 4K UHD Glossy OLED Gaming Monitor
Samsung 27-inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
As an affiliate, we earn on qualifying purchases.
The Deprecation Timeline So Far
The source report says OpenAI lists deprecations on a public page and emails affected customers. It describes an Oct. 23 shutdown covering several older IDs, including the original GPT-4o snapshot, gpt-4o-2024-05-13, as well as gpt-4-turbo, gpt-3.5-turbo, gpt-4.1-nano, o1, o1-pro, o3-mini, o4-mini and gpt-image-1. These dates and model names are reported from the page as checked on Nov. 1, 2026; they should be confirmed against OpenAI’s live schedule.
The report distinguishes the coming API changes from an earlier ChatGPT product change: it says GPT-4o, GPT-4.1 and GPT-4.1 mini were removed from ChatGPT on Feb. 13, 2026. API deprecations affect developers’ integrations, not necessarily the same products or access arrangements used through ChatGPT.
There are other dates for developer tools and models. The source says OpenAI planned to shut down its Evals platform, Agent Builder and reusable prompt objects on Nov. 30, with existing evals read-only from Oct. 31. It also lists later 2027 dates for some audio, speech, transcription and coding models. These are separate items from the December and January language-model deadlines.
““The replacements think before they answer.””
— Thorsten Meyer, ThorstenMeyerAI.com
Migration Details Still Need Checking
The source is a report about OpenAI’s documentation, not a direct reproduction of the full deprecation notice. Readers should check the live OpenAI deprecations page for affected accounts, exact model IDs and any changes made after Nov. 1. The supplied material does not establish whether every customer will receive the same migration options or whether the listed replacement mappings apply to all workloads.
The article also raises a possible notice-period question. It says OpenAI’s page promises at least six months’ notice for generally available models unless safety or compliance concerns require a faster schedule, while the reported notice for the Jan. 14 retirements ran from Oct. 14. The source says the entry does not explain which rule applies. That leaves the basis for the shorter interval unclear; it is not evidence by itself that OpenAI breached a commitment.
Pricing and behavior also require verification. The source gives some per-million-token prices, but the supplied text ends during the comparison and does not include a full table or complete testing results. It does not quantify how often output limits might produce empty answers, or how much prompts would need to change. Those effects will vary by integration.
Audit Model IDs Before Deadlines
Teams should start by searching application code, configuration files, prompt tools, test fixtures and no-code automations for every model ID they use. The Oct. 23 list warrants immediate checks: if a still-active workflow calls one of those IDs, it may already be failing, according to the report. Owners should also identify fine-tuned models, which the source says follow the shutdown schedule of their base model.
For the December and January dates, a measured migration plan should test replacement models against representative prompts before switching production traffic. Check response formats, tool use, latency, token consumption and failure handling; review whether sampling parameters remain supported and whether output limits allow complete responses. Compare costs using current official pricing rather than relying on a partial price excerpt.
The source says OpenAI’s Evals platform is due to close on Nov. 30, with existing evaluations already read-only from Oct. 31. Developers relying on those tools should check the company’s guidance and preserve any test results they need. OpenAI’s deprecation page and account communications remain the key places to confirm dates, access options and any subsequent changes.
Key Questions
When are the next reported OpenAI API model shutdowns?
The source reports that GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro are scheduled for Dec. 11, 2026. GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are listed for Jan. 14, 2027. Confirm the schedule on OpenAI’s current deprecations page.
Could an integration already be affected?
Yes. The report says several model IDs were shut down on Oct. 23, 2026, including gpt-3.5-turbo, gpt-4-turbo, the original GPT-4o snapshot, o1 and o4-mini. Search code and automation settings for those IDs and test any workflow that uses them.
Will changing the model ID be enough?
Not necessarily. The source says replacement reasoning models may handle parameters, output limits, pricing and response behavior differently. Test representative prompts and integrations, including any that depend on a particular format or structured response.
How much will the replacements cost?
The source reports lower listed prices for some GPT-4 family replacements and higher prices for GPT-5 nano and GPT-5 mini replacements. Its supplied pricing comparison is incomplete, so check current OpenAI rates and estimate costs using your own input and output token usage.
Source: ThorstenMeyerAI.com
