Recent Discussions
Trigger cant read fabric data agent
I make an agent in Azure AI Foundry. I use fabric data agent as a knowledge. Everything runs well until I try to use trigger to orchestrate my agent. I have added my trigger identity to fabric workspace where my fabric data agent and my lakehouse located. My trigger can work well and there is no error, but my agent cannot respond as if I do a prompt via the playground. Why?
Azure AI foundry SDK-Tool Approval Not Triggered When Using ConnectedAgentTool() Between Agents
I am building an orchestration workflow in Azure AI Foundry using the Python SDK. Each agent uses tools exposed via an MCP server (deployed in Azure container app), and individual agents work perfectly when run independently — tool approval is triggered, and execution proceeds as expected. I have a main agent which orchestrates the flow of these individual agents.However, when I connect one agent to another using ConnectedAgentTool(), the tool approval flow never occurs, and orchestration halts. All I see is the run status as IN-PROGRESS for some time and then exits. The downstream (child) agent is never invoked. I have tried mcp_tool.set_approval_mode("never") , but that didn't help. Auto-Approval Implementation: I have implemented a polling loop that checks the run status and auto-approves any requires_action events. async def poll_run_until_complete(project_client: AIProjectClient, thread_id: str, run_id: str): """ Polls the run until completion. Auto-approves any tool calls encountered. """ while True: run = await project_client.agents.runs.get(thread_id=thread_id, run_id=run_id) status = getattr(run, "status", None) print(f"[poll] Run {run_id} status: {status}") # Completed states if status in ("succeeded", "failed", "cancelled", "completed"): print(f"[poll] Final run status: {status}") if status == "failed": print("Run last_error:", getattr(run, "last_error", None)) return run # Auto-approve any tool calls if status == "requires_action" and isinstance(getattr(run, "required_action", None), SubmitToolApprovalAction): submit_action = run.required_action.submit_tool_approval tool_calls = getattr(submit_action, "tool_calls", []) or [] if not tool_calls: print("[poll] requires_action but no tool_calls found. Waiting...") else: approvals = [] for tc in tool_calls: print(f"[poll] Auto-approving tool call: {tc.id} name={tc.name} args={tc.arguments}") approvals.append(ToolApproval(tool_call_id=tc.id, approve=True)) if approvals: await project_client.agents.runs.submit_tool_outputs( thread_id=thread_id, run_id=run_id, tool_approvals=approvals ) print("[poll] Submitted tool approvals.") else: # Debug: Inspect run steps if stuck run_steps = [s async for s in project_client.agents.run_steps.list(thread_id=thread_id, run_id=run_id)] if run_steps: for step in run_steps: sid = getattr(step, "id", None) sstatus = getattr(step, "status", None) print(f" step: id={sid} status={sstatus}") step_details = getattr(step, "step_details", None) if step_details: tool_calls = getattr(step_details, "tool_calls", None) if tool_calls: for call in tool_calls: print(f" tool_call id={getattr(call,'id',None)} name={getattr(call,'name',None)} args={getattr(call,'arguments',None)} output={getattr(call,'output',None)}") await asyncio.sleep(1) This code works and auto-approves tool calls for MCP tools. But while using ConnectedAgentTool(), the run never enters requires_action — so no approvals are requested, and the orchestration halts. Environment: azure-ai-agents==1.2.0b4 azure-ai-projects==1.1.0b4 Python: 3.11.13 Auth: DefaultAzureCredential Notes: MCP tools work and trigger approval normally when directly attached. and I ndividual agents function as expected in standalone runs. Can anyone help here..!
AI Foundry - Open API spec tool issue
Hello, I'm trying to invoke my application's API as a tool within the AI Foundry OpenAPI specification tool. However, I keep encountering a 401 Unauthorized error. I'm using a Bearer token for authentication, and it works perfectly when tested via Postman. I'm unsure whether the issue lies in the input/output schema or the connection configuration. Unfortunately, the AI Foundry Traces aren't providing enough detail to pinpoint the exact problem. Additionally, my API and AI Foundry accounts are hosted in different Azure subscriptions and networks. Could this network separation be affecting the connection? I would appreciate any guidance or help to resolve this issue. -TamizhI can't delete my Azure Key Vault Connection in Azure AI Foundry
I have deleted all project under my Azure AI Foundry, but I still can't delete the Azure Key Vault Connection. Error: Azure Key Vault connection [Azure Key Vault Name] cannot be deleted, all credentials will be lost. Why is this happening?
Issue when connecting from SPFX to Entra-enabled Azure AI Foundry resource
We have been successfully connecting our chat bot from an SPFX to a chat completion model in Azure, using key authentication. We have a requirement now to disable key authentication. This is what we've done so far: disabled API authentication in the resource Gave to the SharePoint Client Extensibility Web Application Principal "Cognitive Services OpenAI User", "Cognitive Service User" and "Cognitive Data Reader" permission in the resource In the SPFX we have added the following in the package-solution.json (and we have approved it in the SharePoint admin site): "webApiPermissionRequests": [ { "resource": "Azure Machine Learning Services", "scope": "user_impersonation" } ] To connect to the chat completion API we're using fetchEventSource from '@microsoft/fetch-event-source', so we're getting a Bearer token using AadTokenProviderFactory from "@microsoft/sp-http", e.g.: // preceeded by some code to get the tokenProvider from aadTokenProviderFactory const token = await tokenProvider.getToken('https://ai.azure.com'); const url = "https://my-ai-resource.openai.azure.com/openai/deployments/gpt-4o/chat/completions?api-version=2025-01-01-preview"; await fetchEventSource(url, { method: 'POST', headers: { Accept: 'text/event-stream', 'Content-type': 'application/json', Authorization: `Bearer ${token}` }, body: body, ...// truncated We added the users (let's say, email address removed for privacy reasons) in the resource as an Azure AI User. When we try to get this to work, we get the following error: The principal `email address removed for privacy reasons` lacks the required data action `Microsoft.CognitiveServices/accounts/OpenAI/deployments/chat/completions/action` to perform `POST /openai/deployments/{deployment-id}/chat/completions` operation. How can we make this work? Ideally we would prefer the SPFX principal to do the request to the chat completion API, without needed to have to add end users in the resource thorugh IAC, but my understanding is that AadTokenProviderFactory only issues delegated access tokens.Azure OpenAI: GPT-5-Codex Availability?
Greetings everyone! I just wanted to see if there's any word as to when/if https://openai.com/index/introducing-upgrades-to-codex/ will make it's way to the AI Foundry. It was released on September 15th, 2025, but I have no idea how long Azure tends to follow behind OpenAI's releases. It doesn't really seem like there's any source of information to view whenever new models drop as to what Azure is going to do with them, if any. Any conversation around this would be helpful and appreciated, thanks!Responses API for gpt-4.1 in Europe
Hello everyone! I'm writing here trying to figure out something about the availability of the "responses" APIs in european regions: https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/responses?tabs=python-key i'm trying to deploy a /responses endpoint for the model we are currently using, gpt-4.1, since i've read that the /completions endpoint will be dismissed by OpenAI starting from august 2026. Our application is currently migrating all the API calls from completions to responses, and we were wondering if we could already do the same for our clients in Europe as well, which have to comply to GDPR and currently use our Azure subscription. In the page linked above, i can see some regions that would fit our needs, in particular: francecentral norwayeast polandcentral swedencentral switzerlandnorth but then, i can also read "Not every model is available in the regions supported by the responses API.", which probably already answers my question: from the Azure AI Foundry Portal, i wasn't able to deploy such endpoint in those regions, except for the o3 model. For the 4.1 model, only the completions endpoint is listed, while searching for "Responses" (in "Deploy base model") returns only o3 and these others: Can you confirm that i'm not doing anything wrong (looking in the wrong place to deploy such endpoint), and currently the gpt-4.1 responses API is not available in any European region? Do you think it's realistic it will be soon (like in 2025)? Any european region would work for us, in the "DataZone-Standard" type of distribution, which already ensures GDPR compliance (no need for a "Standard" one in one specific region). Thank you for your attention, have a nice day or evening,Unable to locate and add a VM (GPU family) to my available VM options.
I am using azure AI foundry and need to run GPU workload but N-series VM options do not appear when i try to add quota Only CPU families like D and E are listed How can i enable or request N-series GPU VMs in my subscription and regionAzure Communication Services - Python SDK Call Media not working with CallConnectionClient
Hi team, I’m working on a FastAPI service that uses Azure Communication Services Call Automation (Python SDK) to handle outbound PSTN calls and real-time speech interaction. So far it is able to make phone calls but not able to do media handling part during conversation. Environment: Python version: 3.12-slim Package: azure-communication-callautomation (version: 1.4.0) Hosting: Azure Container Apps speech cognitive resource is connected to azure communication services https://drive.google.com/file/d/1uC2S-LNx_Ybpp1QwOCtqFS9pwA84mK7h/view?usp=drive_link What I’m trying to do: Place an outbound call to a PSTN number Play a greeting (TextSource) when the call is connected Start continuous speech recognition, forward transcript to an AI endpoint, then play the response back Code snippet: # Play greeting try: call_connection = client.get_call_connection(call_id) call_media = call_connection.call_media() call_media.play_to_all( play_source, operation_context="welcome-play" ) print("Played welcome greeting.") except Exception as e: print("Play Greeting Failed: ", str(e)) # start Recognition participants = list(call_connection.list_participants()) for p in participants: if isinstance(p.identifier, PhoneNumberIdentifier): active_participants[call_id] = p.identifier try: call_connection = client.get_call_connection(call_id) call_media = call_connection.call_media() call_media.start_recognizing_media( target_participant=p.identifier, input_type="speech", interrupt_call_media_operation=True, operation_context="speech-recognition" ) print("Started recognition immediately after call connected.") except Exception as e: print("Recognition start failed:", str(e)) break target_participant = active_participants.get(call_id) if not target_participant: print(f"No PSTN participant found for call {call_id}, skipping recognition.") Issue: When the CallConnected event fires,, I get different errors depending on which method I try: 'CallConnectionClient' object has no attribute 'call_media' 'CallConnectionClient' object has no attribute 'get_call_media_operations' 'CallConnectionClient' object has no attribute 'play_to_all' 'CallConnectionClient' object has no attribute 'get_call_media_client' 'CallConnectionClient' object has no attribute 'get_call_media' Also some import errors: ImportError: cannot import name 'PlayOptions' from 'azure.communication.callautomation' ImportError: cannot import name 'RecognizeOptions' from 'azure.communication.callautomation' ImportError: cannot import name 'CallMediaRecognizeOptions' from 'azure.communication.callautomation' ImportError: cannot import name 'CallConnection' ... Did you mean: 'CallConnectionState'? This makes me unsure which API is the correct/updated way to access play_to_all and start_recognizing_media. https://drive.google.com/file/d/1xI-sWil0OKfAfGwjIgG25eD7CEK95rKc/view?usp=drive_link Questions: What is the current supported way to access call media operations (play / speech recognition) in the Python SDK? Are there breaking changes between SDK versions that I should be aware of? Should I upgrade to a specific minimum version to ensure .call_media works? Thanks in advance!
Agent in Azure AI Foundry not able to access SharePoint data via C# (but works in Foundry portal)
Hi Team, I created an agent in Azure AI Foundry and added a knowledge source using the SharePoint tool. When I test the agent inside the Foundry portal, it works correctly; it can read from the SharePoint site and return file names/data. However, when I call the same agent using C# code, it answers normal questions fine, but whenever I ask about the SharePoint data, I get the error: Sorry, something went wrong. Run status: failed I also referred to the official documentation and sample here: https://learn.microsoft.com/en-us/azure/ai-foundry/agents/how-to/tools/sharepoint-samples?pivots=rest I tried the cURL samples as well, and while the agent is created successfully, the run status always comes back as failed. Has anyone faced this issue? Do I need to configure something extra for SharePoint when calling the agent programmatically (like additional permissions or connection binding)? Any help on this would be greatly appreciated. Thanks!Chaining and Streaming with Responses API in Azure
Responses API is an enhancement of the existing Chat Completions API. It is stateful and supports agentic capabilities. As a superset of the Chat Completions class, it continues to support functionality of chat completions. In addition, reasoning models, like GPT-5 result in better model intelligence when compared to Chat Completions. It has input flexibility, supporting a range of input types. It is currently available in the following regions on Azure and can be used with all the models available in the region. The API supports response streaming, chaining and also function calling. In the examples below, we use the gpt-5-nano model for a simple response, a chained response and streaming responses. To get started update the installed openai library. pip install --upgrade openai Simple Message 1) Build the client with the following code from openai import OpenAI client = OpenAI( base_url=endpoint, api_key=api_key, ) 2) The response received is an id which can then be used to retrieve the message. # Non-streaming request resp_id = client.responses.create( model=deployment, input=messages, ) 3) Message is retrieved using the response id from previous step response = client.responses.retrieve(resp_id.id) Chaining For a chained message, an extra step is sharing the context. This is done by sending the response id in the subsequent requests. resp_id = client.responses.create( model=deployment, previous_response_id=resp_id.id, input=[{"role": "user", "content": "Explain this at a level that could be understood by a college freshman"}] ) Streaming A different function call is used for streaming queries. client.responses.stream( model=deployment, input=messages, # structured messages ) In addition, the streaming query response has to be handled appropriately till end of event stream for event in s: # Accumulate only text deltas for clean output if event.type == "response.output_text.delta": delta = event.delta or "" text_out.append(delta) # Echo streaming output to console as it arrives print(delta, end="", flush=True) The code is available in the following github link - https://github.com/arunacarunac/ResponsesAPI Additional details are available in the following links - Azure OpenAI Responses API - Azure OpenAI | Microsoft Learn
Predictions for Artificial Intelligence in next 2-3 years!!!!
2025 - start of agentic AI -Oct 2025: Chatgpt 5 get released (proven to be 10000x times more powerful than chatgpt 4 and can run task automatically) 2026 AI benchmark matches human, beginning of Artificial general intelligence 2027 A new website called letsbuiltai is open source and encourages everyone to train AI. Instead of you training your own AI or an Ai company training their own AI. This would involves everyone training a particular AI simultaneously, paving way for faster and quicker AI growthPush for Rapid AI Growth
There is a key factors of why AI is not growing as quick as speed of light, the reason is because most AI are either built by a specific company (e.g Open AI for chatgpt, Microsoft for Copilot, Google for Gemini). or individuals/small groups building agents for fun or for their workplaces. But what would happen if we merge them together. Imagine, if a website that is own by no one and it is open source and it allows everyone to train the same AI simultaneously at the same time, what would happen. Imagine instead of Microsoft building Copilot, the whole world is building Copilot at the same time, training Copilot at the same time through all global computing power. This would led to an shocking and exponential growth of AI never seen before. This is why I think Copilot should allow everyone to train its AI.Push for hyperrealistic AI Video Generator
I fervently believe that Microsoft must pioneer the development of AI-generated videos. OpenAI has already set the stage with Sora, and if Microsoft doesn't act now, it risks falling behind in the fiercely competitive AI market. This isn't just about keeping pace—it's about leading the charge. Furthermore, the rollout of AI-generated videos must be nothing short of exceptional. These videos need to boast impeccable quality and clearly convey the intended content. Mediocrity has no place in this vision. And let's not forget about preparing Clipchamp for the 2030s. It's imperative to equip it with cutting-edge capabilities that will redefine video creation and editing for the future. Together, these initiatives will not only keep Microsoft at the forefront but will also revolutionize the AI and video landscape.
Azure OpenAI: gpt-5-mini chat/completions streaming returns empty response.
Summary When calling gpt-5-mini via Chat Completions with "stream": true, the server opens the stream but no assistant tokens are emitted and the final JSON is empty (choices: [], created: 0, empty id/model). The same code path streams correctly for gpt-5 and gpt-4o deployments. Also, non-streaming ("stream": false) with gpt-5-mini returns valid content as expected. Environment API: POST /openai/deployments/{deployment}/chat/completions?api-version=2025-01-01-preview Model / Deployment: gpt-5-mini (Azure OpenAI deployment) Date/Time observed: 26 Aug 2025, ~13:00 IST (UTC+05:30) Region: useast2 Note: Same client, headers, and network path work for gpt-5 and gpt-4o streaming. Request Endpoint /openai/deployments/gpt-5/chat/completions?api-version=2025-01-01-preview Body { "messages": [ { "role": "system", "content": "give the best result you can" }, { "role": "user", "content": "Hello" } ], "stream": true } Actual Response (final aggregated JSON after stream ends) { "choices": [], "created": 0, "id": "", "model": "", "object": "", "prompt_filter_results": [ { "prompt_index": 0, "content_filter_results": { "hate": { "filtered": false, "severity": "safe" }, "jailbreak": { "filtered": false, "detected": false }, "self_harm": { "filtered": false, "severity": "safe" }, "sexual": { "filtered": false, "severity": "safe" }, "violence": { "filtered": false, "severity": "safe" } } } ] } Notes: No delta tokens arrive on the SSE stream. No assistant message content is ever emitted. Content filter result is safe across categories. Expected Behavior With "stream": true, server should emit SSE chunks with assistant delta tokens and finish with a populated final message in choices[0].message.content. Azure OpenAI: gpt-5-mini chat/completions streaming returns empty response (choices: [], created: 0) while other models stream fineDo you have experience fine tuning GPS OSS models?
Hi I found this space called Affine. It is a daily reinforcement learning competition and I'm participating in it. One thing that I am looking for collaboration on is with fine tuning GPT OSS models to score well on the evaluations. I am wondering if anyone here is interested in mining? I feel that people here would have some good reinforcement learning tricks. These models are evaluated on a set of RL-environments with validators looking for the model which dominates the Pareto frontier. I'm specifically looking to see any improvements in the coding deduction environment and the new ELR environment they made. I would like to use a GPT OSS model here but its hard to fine-tune these models in GRPO. Here is the information I found on Affine: https://www.reddit.com/r/reinforcementlearning/comments/1mnq6i0/comment/n86sjrk/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_buttonUsing AI to convert unstructured information to structured information
We have a use case to extract the information from various types of documents like Excel, PDF, and Word and convert it into structured information. The data exists in different formats. We started building this use case with AI Builder, and we hit the roadblock and are now exploring ways using the Co-pilot studio. It would be great if someone could point us in the right direction. What should be the right technology stack that we should consider for this use case? Thank you for the pointer.Aidemos Microsoft site doesn't work https://aidemos.microsoft.com/
Hello MS team, I am learning AI-900 in Coursera. The course guides me to try AI demos on https://aidemos.microsoft.com/. But it seems broken for weeks. According to the error message, it could be the issue of the backend. Could the MS team fix it, please? Best Regards, DaleDoc Intelligence: Custom Extraction model | Confidence score deterioration with new formats/layouts
Hi everyone, This is my first time using custom extraction models on the Document Intelligence service, and I would appreciate your input on an experiment I am conducting. I wanted to investigate how these models' confidence scores behave when documents with significantly different format/layout are introduced (later) in the training phase. I started by training models with documents in the same format (some of worse picture quality and slightly rotated), increasingly adding more samples (a new model was trained every time I added new documents, at increments of 5). After every new model was trained, I checked scores against the same, unseen by the model holdout set that had the same format with those in the training set. After training the final model, with 35 identically formatted documents, I started introducing documents with a significantly different format/layout and retraining (at increments of 10). Confidence scores against the holdout set (unchanged) dropped after doing so, without recovering to previous levels. See graph below showing how confidence scores evolved after every training step (adding new documents at every step). Any insights as to why this has happened?
Recent Blogs
- Modern enterprise systems face a simple problem: how to make AI decisions reliable, explainable, and production-ready? This post walks through how the Microsoft Agent Framework structures AI-driven...Oct 10, 2025
- 5 MIN READWhen deploying large language models in Azure AI Foundry, does selecting PTUs (Provisioned Throughput Units) save you money? This is the kind of article that might get its humble writer in hot water,...Oct 09, 2025
