when Azure OpenAI enables or changes its web search tool support
when Azure OpenAI enables or changes its web search tool support? Is there any timeframe which can be shared?
Azure OpenAI Service
Why Azure OpenAI Service still lacks native web search and when will it?
Azure OpenAI Service lacks native web search support that is available on the OpenAI platform, creating a critical feature gap for real-world use cases. Clarification on the rationale and roadmap for this capability would be appreciated.
Azure OpenAI Service
Azure OpenAI Service access approval for Free Trial subscription
I have created an Azure Free Trial subscription and would like to use Azure OpenAI Service for learning and building responsible AI and Agentic AI projects. However, I am unable to see Azure OpenAI Service available in the portal. I understand that…
Azure OpenAI Service
Thread Vector Store Lifecycle
I understand that in AI Foundry, conversations and their associated vector stores expire after 7 days of inactivity. What exactly happens when a VS expires? Does it still occupy memory or storage after expiration, and when is it fully cleaned up from the…
Azure OpenAI Service
Azure OpenAI API returning "No such host is known" once in a while
Issue Summary Intermittent DNS resolution failures when calling Azure OpenAI API from an Azure Web App Service. The API normally works, but after a period of inactivity the first requests occasionally fail with: “No such host is known.” Host:…
Azure OpenAI Service
[Issue] Unified Gating approved: "Sora 2" missing in Model Catalog (East US 2), but "gpt-5-pro" is available
Hello, I received an official confirmation email from Cognitive Services Gating Support approximately 15 hours ago regarding the onboarding to the new Azure OpenAI Service models (including gpt-5-pro, gpt-5.2, and Sora 2). I am writing because I am…
Azure OpenAI Service
Responses API returning deployment name instead of model name
Description I am seeing an inconsistency between the Completions API and the Responses API in Azure OpenAI regarding the value returned in response.model. Deployment name: gpt-5-01 Underlying model: gpt-5 Model version: 2025-08-07 When using the…
Azure OpenAI Service
Error when connection to Azure OpenAI ""AuthenticationtypeDisable" message "Key based authentication is disable for this resource"."
I am connecting to Azure OpenAI but suddently from Dec 30, it shows the error "AuthenticationtypeDisable" message "Key based authentication is disable for this resource". When checking Microsoft Foundry**| Azure OpenAI,** I noticed…
Azure OpenAI Service
Architecture & DevSecOps Patterns for Secure, Multi-tenant AI/LLM Platform on Azure
Hi Microsoft Q&A Team, I am currently designing a production-grade, secure AI/LLM chatbot platform aimed at serving multiple clients (multi-tenant or client-isolated environments). The goal is to establish a robust DevSecOps foundation focusing on…
Azure OpenAI Service
determine max_input_tokens and max_output_tokens
As an agent using Azure OpenAI Service, it only knows the API base, API key, and deployment name. However, the deployment name may not be the same as the underlying model name. The agent often relies on a proxy, such as LiteLLM, to determine…
Azure OpenAI Service
Model retirement - update to what? gpt-4o-transcribe
Dear Team For my endpoint ".../openai/deployments/gpt-4o-transcribe/audio/transcriptions?api-version=2025-03-01-preview" I get this notification: Legacy model version: This deployment is using legacy model gpt-4o-transcribe version 2025-03-20…
Azure OpenAI Service
gpt-4o-transcribe retirment
Legacy model version: This deployment is using legacy model gpt-4o-transcribe version 2025-03-20 which will be retired on 10/15/2025 local time. When the version retires, inferencing will return error responses and all deployments for this version will…
Azure OpenAI Service
AI Observability on Foundry (New vs Classic)
I am trying to implement AI observability (Agent Tracing and Metric Dashboard on Grafana) on Foundry for an agentic RAG project based on Pydantic AI. I have following questions related to this: Foundry classic uses OpenAIInstrumentor whereas New uses…
Azure OpenAI Service
What is the exact use of "Refresh" button under Indexes in Azure Ai Studio?
I am using Azure AI studio to make indexes and I have uploaded 9.6 GB of data in Blob Storage and made a connection of it in Ai Studio, and made a index of it. Later after some days I have 200MB more data that I have uploaded on my Blob. And, updated the…
Azure OpenAI Service
How to delete resource groups in Azure
Hi, I have to delete the resource group created for the Azure Foundry. I currently do not have access to delete it, which is causing hourly costs. Need assist on this, it has been two days like this and this cost is rising up. Regards,
Azure OpenAI Service
Azure OpenAI - 429 - System is experiencing high demand and cannot process your request error
We are currently seeing the exception in our OpenAI resource using gpt 4.o in the Australia East region. Error code: 429 - {"error": {"code": "NoCapacity", "message": "The system is currently experiencing…
Azure OpenAI Service
Text Analysis in PDF Documents with Azure AI Services
Hello everyone, I am working on a project where I regularly handle extensive collections of PDF documents filled with specific technical terms and abbreviations. These documents also vary greatly in layout and formulation, presenting a unique challenge…
Azure AI Search
Azure OpenAI Service
Azure AI Document Intelligence
Azure AI services
Need GPT-5 / GPT-5.1 quota increase applied to correct subscription (EDChartLock, East US 2)
I’m trying to get a GPT-5 / GPT-5.1 quota increase applied to my production subscription, but my portal only shows prioritized Q&A (no option to create a standard support ticket). I’ve already submitted quota requests and emailed Cognitive Services…
Azure OpenAI Service
How do you actually get Azure OpenAI quota increases approved?
Hi there, I'm running a multi-tenant AI platform for European businesses and we've been completely stuck trying to get quota increases for Azure OpenAI for weeks now. We started hitting our GPT-5.1 quota limits constantly a few weeks ago. Due to the low…
Azure OpenAI Service
GPT-4o Throughput Scaling for Production in East US: Cost-Optimized Approach to Achieve 50+ RPM
We’re using the GPT-4o model deployed as Global Standard, with a current throughput limit of 450K tokens per minute. Our application processes one document per request, averaging ~20K input tokens and ~1.5K output tokens (~21.5K tokens total). With this…