🔍 Read the full analysis: OpenAI API Model Shutdown: A Checklist For GPT-4o, GPT-4.1, And GPT-5-nano Users on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
OpenAI’s published API deprecation schedule says several models stopped working on 23 October 2026, with GPT-5 variants scheduled to leave on 11 December and GPT-4o and GPT-4.1 variants on 14 January 2027. Its listed replacements have different prices and reasoning behavior, so developers should check exact model IDs and test their applications before migration.
OpenAI’s API deprecation schedule says GPT-5, GPT-5 mini and GPT-5 nano will be shut down on December 11, 2026, followed by listed GPT-4o and GPT-4.1 snapshots on January 14, 2027. Several other model IDs were removed on October 23, meaning applications still calling those IDs may already be failing.
The schedule, as reported by ThorstenMeyerAI.com after checking OpenAI’s deprecations page on November 1, includes gpt-5, gpt-5-mini and gpt-5-nano, along with gpt-5-pro, o3 and o3-pro, for December 11. OpenAI lists gpt-5.6-sol as the replacement for GPT-5 and o3, gpt-5.6-terra for GPT-5 mini, and gpt-5.6-luna for GPT-5 nano. The listed GPT-4o and GPT-4.1 snapshots are scheduled for January 14, with gpt-6-luna named as their replacement.
Models already removed on October 23 include gpt-4.1-nano, gpt-4-turbo, gpt-3.5-turbo, o1 and o4-mini, as well as the original GPT-4o snapshot. The source report says affected requests can fail if an application still calls a shut-down ID. It also says fine-tuned models follow their base models’ shutdown schedules; the replacements listed in the report do not support fine-tuning.
Migration is not just a matter of changing a model name. The report describes the replacements as reasoning models, with hidden reasoning tokens billed as output. Some sampling parameters may need to be removed, and restrictive output-token limits may produce an empty answer while still incurring a charge. These behavior and cost changes make testing part of the migration, not an optional final check.
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 Applications
The deadlines affect software built on OpenAI’s API, including services, internal tools and automations that may use model IDs buried in configuration files or workflows. If a retired ID remains in production, an application could stop completing the task that depends on it. A change to a replacement model can also alter response formats, latency, output length and billing, even when the application continues to run.
Pricing comparisons in the source report show that costs do not move in one direction. At the listed standard rates for short-context use, GPT-4o is priced at $2.50 per million input tokens and $10 per million output tokens, compared with $0.10 and $0.50 for GPT-6 Luna. By contrast, GPT-5 nano is listed at $0.05 input and $0.40 output, while GPT-5.6 Luna is $0.20 and $1.20. That is four times the input price and three times the output price, according to the report. The figures are a snapshot checked November 1, 2026, not a guarantee of future prices.
For teams, the practical consequence is that model inventories, budgets and quality checks need to be reviewed together. A cheaper replacement may still require prompt changes or new evaluations, while a more expensive one may affect operating costs across high-volume workloads.
Popular right now
Samsung 85-Inch QN70H Neo QLED 4K Smart TV (2026)
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.
OpenAI’s Published Shutdown Calendar
OpenAI’s deprecations page is the source for the dates in the report, which says the company notifies affected customers and documents model removals there. The schedule distinguishes dated model snapshots from short model names. Although the notice lists specific snapshots, the report says the model pages also mark the plain GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini names as deprecated; developers should verify the status of the exact IDs they use.
The same schedule includes shutdowns for related services. The source report says OpenAI planned to close its Evals platform, Agent Builder and reusable prompt objects on November 30, 2026, with existing evaluations becoming read-only on October 31. It also lists later 2027 dates for speech, audio, transcription and other model families. Those items broaden the migration work beyond text-generation endpoints for teams that use them.
OpenAI’s published policy, as summarized by the report, promises at least six months’ notice for generally available models unless safety or compliance concerns require a faster schedule. The January 14 date was announced on October 14, three months earlier. The source says the notice does not explain which policy applies to that deprecation. It also reports that some customers may be able to buy dedicated capacity for continued access after a shutdown.
Open Questions on Dates and Behavior
The source report does not establish whether the three-month notice for the January removals is an exception to OpenAI’s stated six-month notice policy, or why that shorter period applies. Developers with contracts or committed spending are advised in the report to ask their OpenAI account team. The precise eligibility and terms for dedicated capacity after shutdown are also not given.
Migration outcomes remain application-specific. The report identifies possible issues with reasoning-token billing, sampling parameters and output limits, but does not provide test results across individual products or workloads. Pricing may also change after the November 1 check. Teams should confirm current rates and the status of each exact model ID directly with OpenAI before relying on the reported figures.
Check IDs and Test Replacements
Developers should first search production code, configuration, scheduled jobs, test fixtures and no-code workflows for every affected model ID, including dated snapshots and fine-tuned variants. They should then check OpenAI’s current deprecation page and confirm which replacement is listed for each model. The October 23 removals require immediate attention if any affected IDs remain in use.
Before the December 11 and January 14 deadlines, teams can run representative prompts against replacement models in a test environment. The source report recommends checking response quality and format, application handling of reasoning-model outputs, token limits, latency and total input and output costs. Teams should also update or replace evaluation workflows ahead of the reported November 30 closure of Evals and related tools. OpenAI’s published schedule and pricing pages remain the places to verify changes as the dates approach.
Key Questions
When are GPT-5 nano and GPT-5 mini scheduled to leave the API?
OpenAI’s schedule, as reported by ThorstenMeyerAI.com, sets December 11, 2026 for GPT-5 nano and GPT-5 mini, as well as GPT-5 and several other models.
When are the listed GPT-4o and GPT-4.1 models scheduled to be shut down?
The listed GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled for January 14, 2027. The report says the recommended replacement is gpt-6-luna; check OpenAI’s current documentation for the exact model ID and status.
Which affected models have already been removed?
The report says removals on October 23, 2026 included gpt-4.1-nano, gpt-4-turbo, gpt-3.5-turbo, o1, o4-mini and the original GPT-4o snapshot, among others. Calls to those IDs may fail if still present in an application.
Will switching to a replacement model keep the same behavior and cost?
Not necessarily. The reported replacements can have different token prices and reasoning behavior, and may require changes to sampling parameters or output limits. Teams should test their own prompts and calculate costs using current official pricing.
Source: ThorstenMeyerAI.com
