🔍 Read the full analysis: GPT-4o, GPT-4.1, And GPT-5-nano API Retirement: Timing, Costs, And Next Steps 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 published schedule says GPT-5, GPT-5 mini, GPT-5-nano, o3 and related models will leave the API on December 11, 2026; GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled to follow on January 14, 2027. Listed replacements can change token costs and API behavior, so developers should inventory model IDs, verify current pricing and test their applications before switching.
OpenAI plans to retire several API models on December 11, 2026, and January 14, 2027, according to the deprecation schedule reviewed by ThorstenMeyerAI.com on November 1. The dates affect applications using GPT-5, GPT-5-nano, GPT-4o and GPT-4.1 model IDs, with listed replacements that may change both token costs and model behavior.
On December 11, 2026, the schedule lists gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-pro, o3 and o3-pro for retirement. The article says 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. It lists gpt-5.6-sol with pro reasoning mode for gpt-5-pro and o3-pro.
A second group is scheduled for January 14, 2027: GPT-4o snapshots, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots. The source says OpenAI recommends gpt-6-luna for these models. It also reports that a number of IDs, including gpt-4.1-nano, an early GPT-4o snapshot, gpt-4-turbo, gpt-3.5-turbo, o1 and o4-mini, were already shut down on October 23. Calls that still use retired IDs may fail.
The article’s pricing comparison uses standard prices per million tokens for short-context requests, up to 272,000 tokens, and says it was checked November 1. It reports GPT-4o 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 its proposed replacement gpt-5.6-luna is $0.20 and $1.20. Those are the source’s reported rates; developers should check OpenAI’s live pricing before budgeting.
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 Can Change Cost and Behavior
The schedule matters to teams because a model ID is often stored in application code, configuration files, tests or automation tools. If a service continues to call an ID after its shutdown, that request may stop working. Replacing the ID is not necessarily a drop-in change: the source describes the suggested replacements as reasoning models, with different token accounting and request settings.
According to the article, hidden reasoning tokens are billed as output, some sampling parameters must be removed, and a tight output-token limit can result in an empty answer that is still billed. These are claims in the source material, not independently verified here. They underline why teams should test representative requests, output formats, latency and costs rather than treating migration as a text-only edit. The reported prices also show that cost effects differ by model: some listed migrations are cheaper per token, while others are more expensive.
The article also flags related product changes: it says OpenAI plans to shut down its Evals platform, Agent Builder and reusable prompt objects on November 30, 2026. That date comes before the December model retirements and could affect teams’ ability to run or manage parts of their existing testing workflow.
Popular right now
Samsung 85-Inch QN70H Neo QLED 4K Smart TV (2026)
Shark PowerPro Reveal Plus Cordless Vacuum Cleaner, Charcoal, IZ376H
TCL T7 Series 85-Inch 4K 144Hz QLED Smart Google TV
As an affiliate, we earn on qualifying purchases.
The Published Retirement Calendar
The retirement schedule is attributed to OpenAI’s deprecations page, which the source says it checked on November 1, 2026. It reports that OpenAI emails affected customers and documents deprecations on that page. Dates are tied to specific model IDs and snapshots, so teams should compare the schedule with the exact names in their own requests rather than relying only on broad family labels.
The source says fine-tuned models follow the shutdown schedule of their base models, and that the listed replacements do not support fine-tuning. It also reports that ChatGPT access to GPT-4o, GPT-4.1 and GPT-4.1 mini ended on February 13, 2026. That earlier ChatGPT change is separate from the current API schedule: the announced dates concern API use, affecting developers and businesses whose products call the models.
There is also a notice-period question. The article says OpenAI’s page promises at least six months’ notice for generally available models unless safety or compliance concerns require a shorter timeline. It notes that the January 14, 2027 entry was announced October 14, a three-month interval, but says the entry does not explain which notice rule applies. The source reports that some customers may be able to buy dedicated capacity for continued access after a shutdown.
Notice Rules and Migration Effects
The supplied material does not establish why the January 14 date falls three months after its reported October 14 notice despite the stated six-month notice policy. It says the deprecation entry does not identify whether a safety or compliance exception applies. Developers with contracts or committed spend are advised by the source to ask their OpenAI account team about the schedule and any continued-access options.
Actual migration costs will depend on each application’s input and output volume, cached-token use, context length and request pattern. The listed rates are a snapshot from November 1, not a guarantee of future prices. The source also raises behavior changes linked to reasoning models and a revised Model Spec, but provides no independent test results showing how a particular application will respond. Teams will need to verify outputs and API compatibility against their own use cases.
The article reports that the October 29 Model Spec revision removed much of its detailed style guidance, including a default preference for Markdown, and instructs models to interpret instructions by intent. The source does not establish how that revision will affect any specific API model or application. Explicit formatting instructions and regression tests may help identify changes, but the impact remains application-dependent.
Audit IDs Before December
Developers should first search source code, configuration, scheduled jobs, no-code workflows and test fixtures for the exact model IDs on the retirement list. The source recommends treating short names such as gpt-4o and gpt-4.1 as relevant to the January deadline as well as checking dated snapshots. Any ID already retired on October 23 should be investigated immediately.
Before changing production traffic, teams can test the listed replacement with representative prompts and outputs, compare token use and price, and check any sampling parameters or output limits that may need adjustment. They should also test required response formatting explicitly and review any reliance on fine-tuning, Evals, Agent Builder or reusable prompt objects. OpenAI’s published schedule and pricing should be checked again closer to each date because both the source and rates may change.
The next dates in the source’s calendar are the November 30, 2026 shutdown of Evals, Agent Builder and reusable prompt objects; the December 11 GPT-5 and o3 retirements; and the January 14, 2027 GPT-4o and GPT-4.1 retirements. OpenAI’s deprecation page is identified as the place to track notices, while customers with contractual commitments can seek clarification from their account team.
Key Questions
When are GPT-5-nano and GPT-5 mini scheduled to leave the API?
According to the schedule reported by ThorstenMeyerAI.com, gpt-5-nano and gpt-5-mini are listed for retirement on December 11, 2026. The article lists gpt-5.6-luna and gpt-5.6-terra, respectively, as their replacements.
When are GPT-4o and GPT-4.1 API snapshots scheduled to retire?
The source reports that GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled for January 14, 2027. It says OpenAI recommends gpt-6-luna as their replacement.
Will switching to a replacement model always reduce API costs?
No. The source’s November 1 price comparison reports lower listed per-token prices for some migrations, including GPT-4o to gpt-6-luna, but higher prices for others, including gpt-5-nano to gpt-5.6-luna. Actual spending depends on usage and current rates.
What should developers test before migrating?
Check every model ID in code and workflows, then test replacement models with representative requests. Review output quality and format, token use, price, sampling settings, output limits and any reliance on fine-tuning or OpenAI tools scheduled for retirement.
Source: ThorstenMeyerAI.com
