🔍 Read the full analysis: OpenAI’s GPT-4o, GPT-4.1, And GPT-5-nano API Changes: A Migration Guide on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI’s API deprecation schedule, as reported by ThorstenMeyerAI.com on November 1, sets December 11, 2026, retirements for GPT-5 models and January 14, 2027, retirements for GPT-4o and GPT-4.1 snapshots. The report lists newer models as replacements, but flags potential price increases for some users and behavior changes because the replacements use reasoning. Several older model IDs have already been shut down.
OpenAI’s API retirement schedule calls for GPT-5, GPT-5 mini, GPT-5-nano, o3 and related models to be removed on December 11, 2026, followed by GPT-4o and GPT-4.1 snapshots on January 14, 2027, according to a migration guide published by ThorstenMeyerAI.com on November 1. The change affects developers and businesses whose applications call these model IDs; some older IDs listed in the guide were already shut down on October 23.
The guide says OpenAI’s deprecation page lists gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-pro, o3 and o3-pro for retirement on December 11. It identifies gpt-5.6-sol, gpt-5.6-terra and gpt-5.6-luna as replacements, with the mapping varying by model. The guide says GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini snapshots are scheduled to retire on January 14, with gpt-6-luna listed as their replacement.
Among model IDs the guide says had already been shut down on October 23 are gpt-4.1-nano, the original gpt-4o-2024-05-13 snapshot, gpt-4-turbo, gpt-3.5-turbo, o1 and o4-mini. It also lists a November 30 shutdown for OpenAI’s Evals platform, Agent Builder and reusable prompt objects, with existing evals described as read-only since October 31. These dates and mappings are reported by the guide, which says it checked OpenAI’s deprecations page on November 1.
The guide warns that migration may involve more than changing a model name. It describes the listed replacement models as reasoning models, for which hidden reasoning tokens are billed as output. It also says some sampling parameters need to be removed and that a low output-token limit can result in an empty answer that is still billed. OpenAI’s documentation and current pricing should be checked before a migration, as details can change.
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 and Runtime Changes
The schedule matters because a retired model ID can leave an application unable to get responses until its configuration is updated. Model names can be embedded in source code, test fixtures and automated workflows, so teams may need to search beyond the main application settings. The guide says short model names such as gpt-4o and gpt-4.1 should be treated as part of the January retirement, alongside the dated snapshots.
Costs may move in either direction. In the guide’s comparison of standard short-context prices per million tokens, GPT-4o is listed at $2.50 per million input tokens and $10 per million output tokens, while gpt-6-luna is listed at $0.10 and $0.50. By contrast, it lists GPT-5-nano at $0.05 input and $0.40 output, versus $0.20 and $1.20 for gpt-5.6-luna. Those figures are the guide’s November 1 price comparison, not a guarantee of future charges; actual spending depends on usage and applicable pricing.
There is also a behavior and testing risk. A replacement that reasons differently, uses different token accounting, or responds differently to formatting instructions can affect output quality and application costs. Developers should test real workflows—not just whether an API call succeeds—before switching production traffic.
Products Worth Considering
Apple MacBook Pro 14 (Early 2026) – M5 Pro, 15‑Core CPU, 16‑Core GPU, 48 GB Unified Memory, 2 TB SSD, Liquid Retina XDR Display, Silver
Apple MacBook Pro 16.2-inch with M5 Pro, 18-core CPU, 20-core GPU, 24GB Memory, 1TB SSD, Space Black
Apple MacBook Pro 14 2021 with M1 Max, 64GB RAM, 2TB SSD, Silver
As an affiliate, we earn on qualifying purchases.
OpenAI’s Published Retirement Schedule
The source guide says the schedule is drawn from OpenAI’s deprecations page and that OpenAI emails affected customers. It reports that the December retirements were notified on June 11 and the January group on October 14. It also says OpenAI’s notice page promises at least six months’ notice for generally available models unless safety or compliance concerns require a faster timeline. The guide notes that the January group’s listed notice period is three months, but says the entry does not explain which rule applies.
The API changes are distinct from an earlier ChatGPT product change. According to the guide, OpenAI removed GPT-4o, GPT-4.1 and GPT-4.1 mini from ChatGPT on February 13, 2026; this schedule concerns API access used by third-party applications and services. The guide also says fine-tuned versions follow their base model’s shutdown schedule and that the replacement models listed do not support fine-tuning. Customers with committed spend or contracts may want to ask OpenAI about their specific arrangements; the guide reports that some customers may be able to buy dedicated capacity for continued access.
Unresolved Dates, Pricing and Behavior
The guide does not provide a complete account of the API migration process or confirm how every application will behave after switching models. It says OpenAI’s retirement entries do not explain why the January models have a three-month notice period despite the stated six-month policy for generally available models. Whether a particular customer qualifies for continued access through dedicated capacity is also not established.
Prices cited are a snapshot checked on November 1, 2026, and may change. The guide’s warnings about reasoning-token billing, parameter support and empty responses point to possible migration issues, but it does not provide results from a controlled comparison of old and replacement models. Teams will need to verify compatibility, output quality and billing against their own workloads.
Test Before Each Shutdown Date
Organizations using affected model IDs should inventory their API calls and identify which services, scheduled tasks and tests depend on them. The nearest reported milestone is the November 30 shutdown of Evals, Agent Builder and reusable prompt objects; model retirements are scheduled for December 11 and January 14. The guide also lists further audio, text-to-speech and other model retirements during 2027.
Before moving production traffic, teams can test the mapped replacement against representative prompts, verify that supported parameters and output limits are appropriate, and compare token usage and bills. They should also check OpenAI’s current deprecations and pricing pages, since this report reflects information checked on November 1. Customers concerned about notice periods, fine-tuning or continued access can seek clarification from their OpenAI account team.
Key Questions
When are GPT-5 and GPT-4o API models scheduled to retire?
The guide reports December 11, 2026, for GPT-5 models including gpt-5-nano, and January 14, 2027, for listed GPT-4o and GPT-4.1 snapshots. Check OpenAI’s current deprecation page for the latest status.
Which models are listed as replacements?
The guide lists gpt-5.6-sol, gpt-5.6-terra and gpt-5.6-luna for different December retirements. It identifies gpt-6-luna as the replacement for the January GPT-4o and GPT-4.1 group.
Will switching models always reduce API costs?
No. The guide’s November 1 price comparison shows lower listed prices for some migrations and higher prices for others, including GPT-5-nano to gpt-5.6-luna. Prices and total bills depend on the applicable rates and token usage.
What should developers test during migration?
Test API compatibility, supported sampling parameters, output limits, response format and quality, token use, and billing. The guide cautions that reasoning tokens are billed as output for the listed reasoning replacements and that a tight output limit may produce an empty billed response.
Source: ThorstenMeyerAI.com
