Skip to content

Commit d685d01

Browse files
committed
removing section
1 parent 81e0f67 commit d685d01

File tree

1 file changed

+0
-15
lines changed

1 file changed

+0
-15
lines changed

articles/ai-services/agents/concepts/model-region-support.md

Lines changed: 0 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -90,21 +90,6 @@ Model can now be referenced in your code (`Target URI` + `@` + `Model Name`), fo
9090

9191
`Model=https://Phi-4-mejco.eastus.models.ai.azure.com/@Phi-4-mejco`
9292

93-
## Bring Your Own Model
94-
95-
The Azure AI Agent Service also allows you to connect to any chat completion model endpoint that you provide from an external model provider (for example, Perplexity). This option can be used when the desired model is not in the AI Foundry Model Catalog.
96-
97-
The following steps will allow you to utilize any chat-completion model that you have access to:
98-
99-
1. Create a connection to your model provider's chat completion API. Click on the **Connected Resources** page, select **Serverless Connection**, then enter your model’s target URI and key (from your model provider's site).
100-
101-
The Model can now be referenced in your code (`Target URI` + `@` + `Model Name`), for example:
102-
103-
`Model=https://api.perplexity.ai@sonar`
104-
105-
> [!NOTE]
106-
> Using a model that is not natively integrated with the Agents service may result in limited functionality (this will depend on the model you bring). For example, some models will provide all agentic features (for example, Llama Instruct models), but others will only provide basic model interaction via chat. To enjoy the full agentic experience – for example tool calling, and utilization of knowledge sources, using an agent-supporting model in the Foundry Model Catalog is recommended.
107-
10893
## Next steps
10994

11095
[Create a new Agent project](../quickstart.md)

0 commit comments

Comments
 (0)