🔍 Read the full analysis: Could OpenAI’s GPT-4o, GPT-4.1, And GPT-5-nano Switch-Off Raise Your Costs? 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 deprecation schedule says several API models have already stopped working, with GPT-5 family models scheduled for removal on Dec. 11, 2026, and GPT-4o and GPT-4.1 snapshots on Jan. 14, 2027. The source’s comparison of listed token prices shows that some replacements cost less while others cost more, and says reasoning-model behavior may require changes to prompts and output limits.
OpenAI’s API deprecation schedule lists shutdowns for older models including GPT-5, GPT-5 mini and GPT-5-nano on Dec. 11, 2026, followed by GPT-4o and GPT-4.1 snapshots on Jan. 14, 2027. The dates matter to businesses and developers using those model IDs: affected API calls may stop working unless applications are updated, and listed replacements can have different prices and response behavior.
ThorstenMeyerAI.com said it checked OpenAI’s deprecations page on Nov. 1 and reported that a number of listed models had already been shut down on Oct. 23, 2026. These include GPT-3.5-turbo, GPT-4, 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. Applications still calling those IDs may fail, according to the report. OpenAI’s page is the underlying source cited for the schedule.
The next listed API date is Dec. 11, when GPT-5, GPT-5 mini, GPT-5-nano, GPT-5-pro, o3 and o3-pro are scheduled to leave. The report says OpenAI lists GPT-5.6-sol, GPT-5.6-terra and GPT-5.6-luna as replacements for those models. On Jan. 14, 2027, the schedule covers dated GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots, with GPT-6-luna listed as the replacement.
The report also flags a separate Nov. 30 shutdown affecting OpenAI’s Evals platform, Agent Builder and reusable prompt objects. Existing evals reportedly became read-only on Oct. 31. These tools are not models, but their removal could affect teams that use them to test or manage applications during migration.
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 Could Change Bills and Outputs
For API customers, changing a model ID is not necessarily a like-for-like swap. The source’s comparison of OpenAI’s listed standard prices per million tokens finds that the January replacement, GPT-6-luna, has lower listed input and output prices than GPT-4o and GPT-4.1. For example, the report gives GPT-4o’s prices as $2.50 per million input tokens and $10 per million output tokens, compared with $0.10 and $0.50 for GPT-6-luna. GPT-4o mini’s listed rates are closer: $0.15 and $0.60, compared with $0.10 and $0.50.
The December changes could have the opposite effect for some customers. The source reports GPT-5-nano at $0.05 per million input tokens and $0.40 per million output tokens, versus $0.20 and $1.20 for its listed GPT-5.6-luna replacement. It gives GPT-5 mini’s rates as $0.25 and $2.00, compared with $2.00 and $12.00 for GPT-5.6-terra. These are price-list comparisons, not estimates of an individual customer’s total bill: actual costs depend on token use and the applicable pricing terms.
The report also says the listed replacements are reasoning models. It warns that hidden reasoning tokens are billed as output and that some sampling parameters may need to be removed. A constrained output-token limit could also produce an empty answer while still incurring a charge, according to the source. Developers should test both cost and output quality rather than assuming a model swap will preserve existing behavior.
Products Worth Considering
Samsung Essential S3 27-Inch Curved FHD Monitor
Samsung 34-Inch ViewFinity S50GC Ultra-WQHD Monitor
Dell UltraSharp 27 Thunderbolt Hub Monitor (U2726DER)
As an affiliate, we earn on qualifying purchases.
API Changes Follow ChatGPT Removals
The API schedule is distinct from changes to OpenAI’s consumer ChatGPT service. ThorstenMeyerAI.com reports that OpenAI removed GPT-4o, GPT-4.1 and GPT-4.1 mini from ChatGPT on Feb. 13, 2026. The later schedule concerns access through the API, which developers use to build applications, bots and automated workflows.
The report says the schedule names dated model snapshots, while short model names such as gpt-4o and gpt-4.1 may appear separately in code or configuration files. It advises treating those short names as affected when their corresponding snapshots are listed for shutdown. OpenAI’s documentation, as described by the source, says fine-tuned versions follow the shutdown schedule of their base model; the listed replacements do not support fine-tuning.
OpenAI’s Model Spec also changed on Oct. 29, according to the source. The revised document removes much of its detailed style guidance, including a previous default for Markdown formatting, and tells models to interpret instructions by their intent. That is a separate change from API access, but it may matter to developers who rely on consistent output formats without explicitly requesting them.
“The replacements think before they answer.”
— ThorstenMeyerAI.com
Pricing and Notice Questions Remain
The schedule gives dates and replacement names, but the source does not establish how every migration will affect a particular application’s total cost, latency or answer quality. Prices cited are standard per-million-token figures for short-context use, and the report says they were checked on Nov. 1, 2026; pricing may change. Customers need to check current rates and their own token usage before projecting spending.
The report says OpenAI’s documentation promises at least six months’ notice for generally available models, unless safety or compliance concerns require a faster timeline. It notes that the Jan. 14 shutdown notice was dated Oct. 14, a three-month interval, but says the deprecation entry does not explain which exception or policy applies. The source also says some customers may be able to buy dedicated capacity to retain access after a shutdown; eligibility and terms are not specified.
It is also unclear from the supplied material whether every customer has received a notice, what support OpenAI will provide for each migration, or whether all replacement models will be available with the same capabilities and settings. The source attributes the schedule to OpenAI’s documentation, but does not include a direct statement from the company about these open questions.
Audit Model IDs Before Deadlines
Teams using the affected APIs should inventory model IDs across application code, configuration files, tests and no-code workflows, then compare each entry with OpenAI’s current deprecation page. The immediate check is for IDs listed as shut down on Oct. 23; upcoming milestones include the Nov. 30 tools shutdown, the Dec. 11 model removals and the Jan. 14, 2027 GPT-4o and GPT-4.1 snapshot removals.
Before switching production traffic, developers can test the listed replacement against representative prompts, including cases that depend on structured formatting, output length or specific sampling settings. They can also compare token consumption and current published prices, check whether fine-tuning is involved, and verify that applications handle empty or unexpected responses. The source recommends making format requirements explicit in prompts rather than relying on a default style.
OpenAI’s deprecation page and pricing information are the next reference points, and customers with contracts or committed spend can ask their account teams how the schedule applies to them. The supplied material does not give a separate deadline or migration procedure beyond the dates in the published schedule.
Key Questions
When are GPT-5, GPT-5 mini and GPT-5-nano scheduled to leave the API?
They are listed for shutdown on Dec. 11, 2026. The source says OpenAI lists GPT-5.6-sol, GPT-5.6-terra and GPT-5.6-luna as replacements for affected models in that group.
When are GPT-4o and GPT-4.1 API snapshots scheduled to be removed?
The listed GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled for Jan. 14, 2027. The source identifies GPT-6-luna as the listed replacement.
Will switching models necessarily lower API costs?
No. The source’s price comparison shows lower listed rates for some migrations and higher rates for others. For example, it reports higher input and output prices for GPT-5-nano’s listed replacement, GPT-5.6-luna. Actual bills depend on token use and current applicable prices.
What should developers test before migrating?
Check model IDs across code and workflows, then test replacement models with representative prompts. Pay attention to output format, token limits, sampling parameters and billed reasoning tokens, and compare actual usage rather than relying only on per-token prices.
Source: ThorstenMeyerAI.com
