🔍 Read the full analysis: The OpenAI API Switch-Off: Which Apps Break, When, And What Migration Costs on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
OpenAI’s API retirement schedule includes model shutdowns on Dec. 11, 2026, and Jan. 14, 2027; several older IDs stopped working on Oct. 23. The source lists replacement models and prices, but migration may also require changes to prompts, settings and output limits. Exact migration costs depend on usage and are not established by the published per-token prices alone.
OpenAI’s API retirement schedule has already disabled several older model IDs and sets Dec. 11, 2026, and Jan. 14, 2027, as the next major shutdown dates. Developers whose apps call the affected models will need to change model IDs and test their integrations; replacement prices and behavior can differ from what their software currently expects.
According to the deprecation calendar reviewed by ThorstenMeyerAI.com on Nov. 1, models scheduled to leave on Dec. 11 include GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro. The source lists GPT-5.6 Sol, Terra and Luna models as replacements, with o3-pro and GPT-5 pro mapped to GPT-5.6 Sol using its pro reasoning mode.
On Jan. 14, 2027, the schedule lists GPT-4o snapshots, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots for retirement. The recommended replacement given for those models is GPT-6 Luna. The source says earlier retirements on Oct. 23 included GPT-3.5 Turbo, GPT-4 Turbo, GPT-4.1 nano, the original GPT-4o snapshot, o1, o1-pro, o3-mini, o4-mini and the image model GPT-image-1. Calls using those IDs may already fail.
The cost comparison in the source uses OpenAI’s listed standard prices per million tokens for prompts up to 272,000 tokens, checked Nov. 1. For example, it reports GPT-4o at $2.50 per million input tokens and $10 per million output tokens, compared with GPT-6 Luna at $0.10 and $0.50. But GPT-5 nano’s listed replacement, GPT-5.6 Luna, costs $0.20 input and $1.20 output, compared with $0.05 and $0.40. These are token-price comparisons, not estimates of a particular app’s total migration bill.
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 Costs Beyond Token Rates
The changes affect more than a configuration string. The source describes the replacements as reasoning models, which may handle prompts differently and may bill hidden reasoning tokens as output. It also says some sampling parameters must be removed, and that a restrictive output-token limit can produce an empty response that is still billed. Developers need to validate both application behavior and operating costs before switching production traffic.
For teams using older GPT-4-family models, listed per-token prices for GPT-6 Luna are lower in the source’s comparison. The GPT-5 nano and GPT-5 mini replacement examples go the other way, with higher listed rates. Actual spending will depend on input and output volume, caching, reasoning-token use, and any changes in how much text the replacement generates. The material provides no usage data or migration labor estimate, so it cannot establish a universal total cost.
The API deadlines matter to businesses operating customer-support bots, internal tools, automations and other products built on these model IDs. A missed dependency can cause requests to fail; a poorly tested change can produce altered formats or answers even when requests continue to run.
Products Worth Considering
LG UltraGear 32GX850A-B 32-Inch 4K UHD Glossy OLED Gaming Monitor
Samsung 27-inch Odyssey OLED G6 QD-OLED QHD Gaming Monitor
Samsung 27-Inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
As an affiliate, we earn on qualifying purchases.
The API Dates and Earlier Changes
The schedule is separate from a ChatGPT product change. The source reports that OpenAI removed GPT-4o, GPT-4.1 and GPT-4.1 mini from ChatGPT on Feb. 13, 2026; the current dates concern models accessed through the API and therefore affect third-party apps and services.
The source also flags 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. Later dates in the listed schedule include text-to-speech retirement on Jan. 6, 2027; older audio and realtime models on Jan. 20; Whisper and GPT-4o transcription models on Feb. 26; and GPT-5.1, GPT-5.4 nano and GPT-5.3 Codex on April 1.
OpenAI’s deprecation page is the source for the dates and model mappings in the report. The article says OpenAI promises at least six months’ notice for generally available models unless safety or compliance concerns require a shorter timeline. It notes that the Jan. 14 notice was posted Oct. 14, but does not explain whether an exception applies.
“The replacements think before they answer.”
— ThorstenMeyerAI.com, summarizing the replacement-model changes
Unanswered Notice and Cost Questions
The supplied material does not establish how many applications or customers use the affected model IDs, nor does it quantify the engineering time or total cost of migration. Listed token prices alone cannot show whether an individual app’s bill will rise or fall after accounting for usage and reasoning tokens.
It is also unclear why the January retirement notice appears to provide three months before the deadline despite the stated six-month notice policy for generally available models. The source says the deprecation entry does not identify which rule applies. OpenAI says some customers may be able to buy dedicated capacity for continued access after a shutdown, but eligibility, availability and pricing are not provided. Model behavior can vary by application, and the source gives no independent test results comparing replacements with the retiring models.
Audit, Test and Confirm Access
Teams can start by searching application code, environment settings, prompt templates, test fixtures and no-code automations for every affected model ID, including dated snapshots and short names. The source recommends testing replacements before deadlines; a practical migration check should compare representative inputs, required response formats, tool calls, latency and token use, then review results before routing live traffic.
Developers should also review API parameters for settings the replacement models do not support and check output limits that could truncate or suppress answers. Because the source’s price figures were checked on Nov. 1, teams should consult OpenAI’s current pricing and deprecation pages before budgeting or making deployment decisions. Customers with contractual commitments or a need for continued access can ask their OpenAI account team about the notice period and dedicated capacity.
The next listed milestones are the Nov. 30 tooling shutdown, the Dec. 11 model retirements and the Jan. 14 GPT-4-family retirements. Later audio and model changes are scheduled into April 2027, according to the source’s review of OpenAI’s calendar.
Key Questions
Which OpenAI API models are scheduled to shut down next?
The source lists GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 pro, o3 and o3-pro for Dec. 11, 2026, followed by specified GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots on Jan. 14, 2027.
Have any of the listed models already stopped working?
Yes. The source says several IDs, including GPT-3.5 Turbo, GPT-4 Turbo, GPT-4.1 nano, o1 and o4-mini, were shut down on Oct. 23, 2026. Requests that still use a retired ID may fail.
Will switching models reduce API costs?
Not in every case. The source’s listed prices show lower per-token rates for some GPT-4-family replacements, but higher rates for the cited GPT-5 nano and GPT-5 mini replacements. Actual bills depend on usage, output and reasoning tokens, and current pricing.
What could require code changes besides the model ID?
The source says replacement reasoning models may require removing unsupported sampling parameters and revisiting output limits. Teams should also test response formats and application behavior; the source provides no guarantee that replacements will respond identically.
Does the schedule affect ChatGPT users directly?
The dates discussed here concern the API, which supports apps and services built by developers. The source separately reports earlier ChatGPT model removals on Feb. 13, 2026.
Source: ThorstenMeyerAI.com
