issue faced when trying to finetune openai gpt3.5-turbo model with api_base

Question

I am trying to finetune gpt3.5-turbo model

when using

import openai
import os

openai.api_key = os.getenv("OPENAI_API_KEY")
tester = openai.File.create(file=open("testdata.jsonl",'rb'), purpose="fine-tune")

it works perfectly. I can proceed and get a finetuned model

But when working with "Azure OpenAI"

import openai
import os

openai.api_base = "https://tester.openai.azure.com"
openai.api_version = "2023-05-15"
openai.api_key = os.getenv("OPENAI_TESTER_API_KEY")

tester = openai.File.create(file=open("testdata.jsonl",'rb'), purpose="fine-tune")

I get an error

Answer 1

@Maswadkar Currently, fine tuning is turned off for new customers due to capacity constraints while using an Azure OpenAI resource. You can look up the models page of Azure OpenAI documentation where the models have listed fine tuning as not available(NA).

As more regions are added to Azure OpenAI, this restriction might be removed and fine tuning will be available to new users. I hope this helps!! Thanks!!

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.