OpenAI
ft-davinci-002 model retirement
Stopped workingSubscribe to OpenAIShare
Calls to the model ft-davinci-002 on OpenAI stopped working on October 23, 2026.
The error you get
What OpenAI answers a call to ft-davinci-002 once the model is gone, as recorded from real responses:
{"error":{"message":"The model `ft-davinci-002` has been deprecated, learn more here: https://platform.openai.com/docs/deprecations","type":"invalid_request_error","param":null,"code":"model_not_found"}}What to use instead of ft-davinci-002
Move to gpt-5.6-terra, recommended by OpenAI.
Facts
- API
- OpenAI
api.openai.com - Model
ft-davinci-002- Kind
- Model retirement
- Takes effect
- October 23, 2026
- Calls after that day
- Fail
- Source
- Deprecations page
- Last updated
- Verified
- , against OpenAI's own pages
Questions
When does ft-davinci-002 shut down?
Calls to the model ft-davinci-002 on OpenAI stopped working on October 23, 2026.
What happens to calls after October 23, 2026?
They have failed since October 23, 2026.
What error does ft-davinci-002 return once it is shut down?
OpenAI answers HTTP 404: "The model `ft-davinci-002` has been deprecated, learn more here: https://platform.openai.com/docs/deprecations". Send another model to fix it.
How do I find the calls this hits in my code?
Okapi reads your repository, finds the calls this change hits and opens a pull request with the fix. It never commits to your branches: every change comes as a pull request you review.
More from OpenAI 10
-
ft-babbage-002Model retirement Stopped working -
ft-gpt-3.5-turboModel retirement Stopped working -
ft-gpt-4Model retirement Stopped working -
ft-gpt-4.1-nanoModel retirement Stopped working -
ft-gpt-4.1-nano-2025-04-14Model retirement Stopped working -
ft-o4-miniModel retirement Stopped working -
ft-o4-mini-2025-04-16Model retirement Stopped working -
gpt-3.5-turboModel retirement Stopped working -
gpt-3.5-turbo-0125Model retirement Stopped working -
gpt-3.5-turbo-completionsModel retirement Stopped working
You fixed this one by hand.
Okapi opens the pull request with the fix before the next breaking change reaches your code.