openAI bad request error

Question

openAI bad request error

Anonymous

openai.BadRequestError: Error code: 400 - {'error': {'code': 'invalidPayload', 'message': 'The specified base model does not support fine-tuning.'}}

i have selected gpt-4o-mini which can be fine tuned why am i getting

0 comments

Answer accepted by question author

0 additional answers

Your answer

Answer 1

Sina Salam 30,166 Volunteer Moderator

Hello Menon, Kousthub R,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand you're encountering a 400 - invalidPayload error while trying to fine-tune the gpt-4o-mini model using OpenAI's API.

This issue typically is arising not because the model is unsupported, but due to using an incorrect model ID.

Below are supported models for finetuning in Azure OpenAI.

Model ID	Standard Training Regions	Global Training (Preview)	Max request (tokens)	Training Data (up to)	Modality
		-
`gpt-35-turbo` (1106)	East US2 North Central US Sweden Central Switzerland West	-	Input: 16,385 Output: 4,096	Sep 2021	Text to Text
`gpt-35-turbo` (0125)	East US2 North Central US Sweden Central Switzerland West	-	16,385	Sep 2021	Text to Text
`gpt-4o-mini` (2024-07-18)	North Central US Sweden Central	-	Input: 128,000 Output: 16,384 Training example context length: 65,536	Oct 2023	Text to Text
`gpt-4o` (2024-08-06)	East US2 North Central US Sweden Central	-	Input: 128,000 Output: 16,384 Training example context length: 65,536	Oct 2023	Text & Vision to Text
`gpt-4.1` (2025-04-14)	North Central US Sweden Central	✅	Input: 128,000 Output: 16,384 Training example context length: 65,536	May 2024	Text & Vision to Text
`gpt-4.1-mini` (2025-04-14)	North Central US Sweden Central	✅	Input: 128,000 Output: 16,384 Training example context length: 65,536	May 2024	Text to Text
`gpt-4.1-nano` (2025-04-14)	North Central US Sweden Central	✅	Input: 128,000 Output: 16,384 Training example context length: 32,768	May 2024	Text to Text
`o4-mini` (2025-04-16)	East US2 Sweden Central	-	Input: 128,000 Output: 16,384 Training example context length: 65,536	May 2024	Text to Te

Moreover, fine-tuning is only available to users on a paid plan, and your training file must be uploaded in .jsonl format via the Files API. If any of these conditions are unmet, the API will reject your request with the error you're seeing.

For full fine-tuning documentation visit Azure OpenAI Fine-tuning Guide as listed below:

https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models?tabs=global-standard%2Cstandard-chat-completions#fine-tuning-models

I hope this is helpful! Do not hesitate to let me know if you have any other questions or clarifications.

Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.

Leo Tran 5 Reputation points

2025-05-23T10:00:24.35+00:00

Thank you, Sina Salam, for sharing the resolution—your insight is spot on.

Hi Menon, Kousthub R, please refer to Sina Salam's solution below. Should you need any additional clarification, feel free to reach out. If the issue persists, kindly submit a support ticket so we can investigate further.
Anonymous

2025-05-23T10:08:14.79+00:00

@Sina Salam Thank you so much for taking your time and answering my question i have given the right model ID does the region matter im in east us and the chart u sent this particular region was not mentioned so then how can i resolve this issue as my model ID is right and im giving the training dataset in jsonl format
Sina Salam 30,166 Reputation points Volunteer Moderator

2025-05-25T21:18:27.89+00:00
Hello Menon, Kousthub R,

Thank you for your feedback and more clarification that you're using the correct model ID and .jsonl format.

Yes, region plays a critical role in fine-tuning availability. Based on the current Azure OpenAI documentation, the gpt-4o-mini (2024-07-18) model is only supported for fine-tuning in North Central US and Sweden Central regions. Since you're operating in East US, this is likely the root cause of the 400 - invalidPayload error you're having.

Your next steps to resolve:

Switch to a supported region (e.g., North Central US or Sweden Central) for your fine-tuning job.

Alternatively, consider using a different model that is supported in East US, such as:

gpt-35-turbo (1106)

gpt-35-turbo (0125)

o4-mini (2025-04-16)

You can find the full list of supported models and regions in the Azure OpenAI Fine-tuning Guide.

If switching regions isn't feasible, I recommend submitting a support ticket to explore potential workarounds or future availability in East US.

Please, don't forget to close this thread by upvoting and accept as an answer for other community members.
Anonymous

2025-05-26T06:04:07.0033333+00:00

Thank you so much @Sina Salam for your valuable time and giving me this clarity
Owen Burns 0 Reputation points

2025-11-14T22:22:01.68+00:00
I am having the same issue, but I am specifying the full model name correctly. I am trying to use "gpt-4o-2024-08-06", and when I go to the foundry resource in the azure portal I can create a fine tuning job without any problem, but when I try to do the exact same thing in code it gives me this error.

ft_job = await self.client.fine_tuning.jobs.create( training_file=training_file_id, validation_file=validation_file_id, seed=105, model="gpt-4o-2024-08-06" )

Share via

openAI bad request error

0 additional answers

Your answer