Skip to content

Commit 35ecc9c

Browse files
authored
Merge pull request #108 from mark-szabo/patch-1
Revise installation steps for Azure OpenAI integration
2 parents 141c53d + 912b94e commit 35ecc9c

1 file changed

Lines changed: 15 additions & 15 deletions

File tree

README.md

Lines changed: 15 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -38,23 +38,23 @@ This is equivalent to the built-in [OpenAI Conversation integration](https://www
3838

3939
</center>
4040

41-
4241
# Installation and Configuration
4342

44-
1. Download and install the integration from HACS: [Azure OpenAI Conversation](https://my.home-assistant.io/redirect/hacs_repository/?owner=joselcaguilar&repository=azure-openai-ha&category=integration).
45-
2. Restart your Home Assistant instance
46-
3. Go to [Settings -> Devices & Services -> Add Integration -> Azure OpenAI Conversation](https://my.home-assistant.io/redirect/config_flow_start/?domain=azure_openai_conversation)
47-
4. To have a conversation, made sure to deploy a chat completion model (like gpt-4o-mini or gpt-4.1-mini) in Azure. If your model is not the default `gpt-4o-mini`, then configure it on step 6.
48-
5. If you want to generate images using the available `generate_image` service, make sure to deploy the `dall-e-3` model as well.
49-
5. Type your `API Key`, `API Base` used following the example below and hit submit:
50-
```yaml
51-
API Key: 1234567890abcdef1234567890abcdef
52-
API Base: https://your-resource.openai.azure.com/
53-
54-
#API Version is required only for versions prior 3.1.0
55-
API Version: 2025-03-01-preview
56-
```
57-
6. Configure your assistant to use the Azure OpenAI Conversation.
43+
## Configuring Azure Models
44+
45+
1. Deploy an [Azure AI Foundry](https://portal.azure.com/#create/Microsoft.CognitiveServicesAIFoundry) instance to a **region supported by the [Responses API](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/responses?tabs=python-secure#region-availability)**.
46+
*(If you already have a Foundry instance, you can skip this step.)*
47+
2. To enable conversations, deploy a chat completion model (such as `gpt-4o-mini` or `gpt-4.1-mini`).
48+
*If your model is not the default `gpt-4o-mini`, you’ll need to configure it later in step 6.*
49+
3. If you want to generate images using the `generate_image` service, also deploy the `dall-e-3` model.
50+
51+
## Setting Up the Integration
52+
53+
4. Download and install the integration from HACS: [Azure OpenAI Conversation](https://my.home-assistant.io/redirect/hacs_repository/?owner=joselcaguilar&repository=azure-openai-ha&category=integration).
54+
5. Restart your Home Assistant instance.
55+
6. [Click here](https://my.home-assistant.io/redirect/config_flow_start/?domain=azure_openai_conversation) or go to **Settings → Devices & Services → Add Integration → Azure OpenAI Conversation**.
56+
7. Enter your `API Key` and `API Base URL` (use the format `https://your-resource.openai.azure.com/`) and hit **Submit**.
57+
8. Configure your assistant to use the Azure OpenAI Conversation.
5858

5959
# Options
6060

0 commit comments

Comments
 (0)