How to use fine-tuned LLMs from an account but billing for another account
Hi there, I have completed GPT finetuning (Azure openAI) and got fine-tuned LLM models though one my Azure account but which reaches its expense limit, so I want to use my another Azure account to continue work on these fine-tuned models (e.g.,…
Azure OpenAI Service
Azure OpenAI text-embedding-ada-002 deployment returning 503 status codes
I have been using a text-embedding-ada-002 deployment on my Azure OpenAI endpoint to embed large amounts of texts. This worked perfectly until some time last week, when the service started to at sometimes return 503 status codes with message "The…
Azure OpenAI Service
GPT-4.1 Region Availability
I see that gpt-4.1 (along with it's mini and nano versions) is only available in specific regions. Is there a timeline to expand the region availability of these models? Can we expect to see them at the availability of 4o and 4o-mini before those models…
Azure OpenAI Service
Does Azure OpenAI support PDF upload for GPT 4o?
I see this forum that PDFs are supported natively in GPT4o https://community.openai.com/t/direct-pdf-file-input-now-supported-in-the-api/1146647 Are PDFs also available in Azure OpenAI? If not, when will they be available?
Azure OpenAI Service
AzureOpenAI API Calls are crashing
GPT 4o deployment is not responding and producing the error - Validation error at #/data_sources/0/azure_search/parameters/endpoint: Field required
Azure OpenAI Service
Request to Enable Azure OpenAI On Your Data Preview for Azure SQL Server Data Sources
Please enable Azure OpenAI On Your Data preview for Azure SQL Server data sources on subscription "f1d66b39-fc51-4372-923e-baac794f6d76"
Azure OpenAI Service
GPT4.5 stopped processing Base64 image?
Hi. What worked fine until a few days ago has stopped working now. We use GPT4.5 to process Base 64 images. model name: gpt-4.5-preview api_version: 2024-12-01-preview { "type": "image_url", "image_url": { …
Azure OpenAI Service
gpt4o-mini deprecation and replacement in EU
In this documentation, you indicate that the gpt-4o-mini model will be deprecated on August 16, 2025. However, this change is not mentioned in the Azure OpenAI “What’s New” page. Could you please confirm if this deprecation is indeed…
Azure OpenAI Service
How do I fix the ''Model {modelName} is enabled only for api versions 2024-12-01-preview and later" Error for o3-mini model in AzureAI?
We are getting the following error when trying to use the o3-mini in a prompt flow: 'Error Code 400 - {'error': {'code': 'BadRequest','message': 'Model {modelName} is enabled only for api version2024-12-01-preview and later'}} When I make a call to the…
Azure OpenAI Service
Model update question
Hi, I'm a bit confused about the model updates and have two questions: If we choose automatic updates, does the model update itself without any customer action required—for example, like gpt-35-turbo-16k and gpt-4-32k? Can the customer continue using…
Azure OpenAI Service

How to Feed Large, Time-Varying Context to Server-Side LLMs for Summarization and Managing Token Limits When Sending Evolving Context to LLMs
We have a use case where data from one of our APIs evolves over a two-day period. This data is shown to customers on the UI, and we want to pass the same data to a Copilot model so users can ask questions or request summaries based on that…
Azure OpenAI Service
How to implement the AI service for a Shopify store?
Hello, we're interested in the chat service on Microsoft store, it's incredible. we'd like to make a silmar one with our Shopify store. We want the AI service to help the customer to find product, place order, refund, track order, etc. and also can get…
Azure OpenAI Service
404 reply when recursively listing input items with Responses API
When listing input items with the Responses API we get the first 20 items. When we try to list the next 20 items using last_id and the after parameter we get an error message like this: NotFoundError: 404 No item found with id…
Azure OpenAI Service
Is it possible to directly connect Azure OpenAI to Azure Databricks tables?
I am hoping to develop a chatbot application with the Azure OpenAI service in Azure Databricks. To do so, I would like the gpt model to have access to the Azure Databricks tables. Rather than making the LLM generate or modify a SQL query to indirectly…
Azure OpenAI Service

Azure OpenAI Batch Jobs are getting stuck in Validating stage
There are so many Azure Open AI batch jobs stuck in validating stage starting from April 11, 7.16 AM PDT and there isn't a single batch reached completion. We had raised a similar issue on March 18 and was resolved by using a new deployment model, but…
Azure OpenAI Service


Does Azure OpenAI Service Allow Models to Access the Internet?
Given the significant advancements in Azure OpenAI Service, particularly with models like GPT-4, GPT-4 Turbo, and the newly introduced GPT-4o, can any of these models natively access the internet or perform real-time web searches as part of their…
Azure OpenAI Service
Azure AI Document Intelligence
Hidden Throttling Issue with o3 Models in AsyncAzureOpenAI Client
Experiencing what appears to be throttling (not rate limiting) when making parallel calls using the AsyncAzureOpenAI client with o3 models. The requests take longer to complete even when executed in parallel. Context: Granted a quota of 10M TPM for…
Azure OpenAI Service
Chat completion endpoint with vision model returning invalid image data
When we pass base64 image data to image_url, it is throwing invalid image data error. But the image data seems to be valid as we are able to create the image using the data. Below is the request url and request body is attached. please suggest. POST…
Azure OpenAI Service
Got Error in chat completion API: At most 1 image(s) may be provided in one request
Description: Got Error when send two images to LLM Llama-4-Maverick-17B-128E-Instruct-FP8, is it because this model's API does not support multiple images in a single request? it's working fine if only one image. Error: raise…
Azure OpenAI Service
Assistant API Stream in gpt-4o (2024-11-20) mis-encodes function call arguments
I’m using the Azure OpenAI Assistants Playground with gpt-4o model version 2024-11-20. After adding a simple function (like the one below) and then asking “今日のニュースを教えて?” in the Playground, the function call’s arguments appear garbled in the JSON…