BYOM: Using Azure AI Foundry models in Copilot Studio
Copilot Studio gives you a fast, secure way to build conversational agents and deploy them into the Microsoft 365 environment without writing code. But what happens when the default models aren’t enough? That’s where Azure AI Foundry comes in. Copilot Studio now supports direct integration with models deployed via AI Foundry. That means you can choose exactly the model you need, like a language model with stronger reasoning capabilities, an image analysis model, or something domain-specific, and then connect it directly to your Copilot Studio agent.
Example
An IT team is managing support requests across multiple business units. Users often submit screenshots: Error messages, UI glitches, device warnings, but rarely with context. Traditional agents struggle with this format. Text-only prompts don’t help when the issue is visual. Instead of trying to route these through a human queue, we use Azure AI Foundry to deploy a vision model capable of:
- Performing OCR on screenshots
- Identifying key UI elements
- Classifying common issues based on screen layouts
We connect the model to a Copilot agent in Microsoft Teams. Now, users can upload a screenshot, and the agent can recognize a known error and suggest next steps within seconds. No manual ticket triage, no copy-paste from Outlook to ServiceNow.
How to connect an AI Foundry model to your Copilot agent
- Find a model in the huge AI Foundry catalog. You can also fine-tune an existing model
- Deploy your model in Azure AI Foundry
- You will get 3 values, that you need in one of the following
- A deployment name
- An endpoint URL
- An API key
Once you have that, navigate to Copilot Studio and open the agent you want to enhance.
- Go to Settings –> Generative AI
- In the Primary response model dropdown, select Connect a model in the From Azure AI Foundry (preview) section
Once prompted, paste in the:
- Target URI (endpoint)
- Deployment name
- API key
Copilot Studio will verify the connection. Once successful, your agent is now powered by that specific model.
Boom, done! 🧨
PS: This is a super new feature. In case you don’t see the From Azure AI Foundry (preview) section please be patient, it will roll out soon™️. Also make sure, that in your Power Platform environment, you have the Get new features early switch set to Yes

Published on:
Learn moreRelated posts
You are holding GitHub Copilot Wrong!
Most developers think that one can’t really use GitHub Copilot wrong. There is a chat interface that lets you also choose a model, so th...
You are holding GitHub Copilot Wrong!
Part 0 showed why constant prompting, re-prompting, and steering GitHub Copilot feels fragile. Not because Copilot is unreliable, but because ...
Building a Multi-Hierarchy Ticket Classification System (Because Keywords Aren't Enough)
Look, I love a good keyword-based system as much as the next developer. They’re fast, predictable, and when your user says “VPN,&r...
Secretless cross-tenant dataverse access
Client secrets are like hiding your house key under the mat; easy to grab and impossible to audit. Certificates are just slightly better, beca...
How Azure CLI handles your tokens and what you might be ignoring
Running az login feels like magic. A browser pops up, you pick an account, and from then on, everything just works. No more passwords, no more...
How Dev Proxy teaches you to make your apps more resilient
I added Microsoft Dev Proxy to my Mermaid → Dataverse converter, because I wanted to test how it handled rate limits and API errors. What I go...
Building Azure functions that never store secrets — ever
What if your function could hit Microsoft Graph with no client secrets, no certs, and no Key Vault entries? That is exactly what a Managed id...
Introducing Mermaid to Dataverse Converter
Why diagrams matter (and why they usually fail us) Entity-Relationship Diagrams (ERDs) are the universal shorthand for talking about data mode...
It’s OK to be seen trying
It’s OK to be seen trying Somewhere along the way, we started believing that you’re only allowed to speak after you’ve figured everything out....
Stuck in pilot - Part 1: no foundations, no future
“We just need to test the AI. We’ll figure out the data later.” That sentence has quietly killed more AI pilots than any model failure ever ...