Merge pull request #4201 from MicrosoftDocs/main

4/17/2025 AM Publish

Author: Taojunshen

articles/ai-foundry/concepts/models-featured.md

@@ -209,10 +209,12 @@ For more examples of how to use Meta Llama models, see the following examples:
 
 ## Microsoft
 
-Phi is a family of lightweight, state-of-the-art open models. These models were trained with Phi-3 datasets. The datasets include both synthetic data and the filtered, publicly available websites data, with a focus on high quality and reasoning-dense properties. The models underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
+Microsoft models include various model groups such as MAI models, Phi models, healthcare AI models, and more. To see all the available Microsoft models, view [the Microsoft model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
+
 
 | Model  | Type | Capabilities |
 | ------ | ---- | ------------ |
+| [MAI-DS-R1](https://ai.azure.com/explore/models/MAI-DS-R1/version/1/registry/azureml) | [chat-completion with reasoning content](../model-inference/how-to/use-chat-reasoning.md?context=/azure/ai-foundry/context/context) | - **Input:** text (163,840 tokens) <br /> - **Output:** text (163,840 tokens) <br />  - **Tool calling:** No <br /> - **Response formats:** Text. |
 | [Phi-4-multimodal-instruct](https://ai.azure.com/explore/models/Phi-4-multimodal-instruct/version/1/registry/azureml) | [chat-completion (with image and audio content)](../model-inference/how-to/use-chat-multi-modal.md?context=/azure/ai-foundry/context/context) | - **Input:** text, images, and audio (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Tool calling:** No <br /> - **Response formats:** Text |
 | [Phi-4-mini-instruct](https://ai.azure.com/explore/models/Phi-4-mini-instruct/version/1/registry/azureml) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) | - **Input:** text (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 | [Phi-4](https://ai.azure.com/explore/models/Phi-4/version/2/registry/azureml) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) | - **Input:** text (16,384 tokens) <br /> - **Output:** text (16,384 tokens) <br /> - **Tool calling:** No <br /> - **Response formats:** Text |
@@ -226,12 +228,9 @@ Phi is a family of lightweight, state-of-the-art open models. These models were
 | [Phi-3-medium-128k-instruct](https://ai.azure.com/explore/models/Phi-3-medium-128k-instruct/version/6/registry/azureml) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) | - **Input:** text (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 | [Phi-3-medium-4k-instruct](https://ai.azure.com/explore/models/Phi-3-medium-4k-instruct/version/5/registry/azureml) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) | - **Input:** text (4,096 tokens) <br /> - **Output:** text (4,096 tokens) <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 
+#### Inference examples: Microsoft models
 
-See [this model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
-
-#### Inference examples: Microsoft Phi
-
-For more examples of how to use Phi-3 family models, see the following examples:    
+For more examples of how to use Microsoft models, see the following examples:    
 
 | Description                               | Language          | Sample                                                          |    
 |-------------------------------------------|-------------------|-----------------------------------------------------------------|    
@@ -242,6 +241,9 @@ For more examples of how to use Phi-3 family models, see the following examples:
 | Llama-Index                               | Python            | [Link](https://aka.ms/azureai/llamaindex)             |  
 
 
+See [the Microsoft model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
+
+
 ## Mistral AI
 
 Mistral AI offers two categories of models, namely: 

articles/ai-foundry/includes/region-availability-maas.md

@@ -73,10 +73,11 @@ Llama 3.1 8B Instruct <br> Llama 3.1 70B Instruct <br> Llama-3.2-11B-Vision-Inst
 Llama 3.1 405B Instruct  | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)    | East US <br> East US 2 <br> North Central US <br> South Central US <br> West US <br> West US 3     | Not available  |
 
 
-### Microsoft Phi family models
+### Microsoft models
 
 | Model | Offer Availability Region  | Hub/Project Region for Deployment  | Hub/Project Region for Fine tuning  |
 |---------|---------|---------|---------|
+MAI-DS-R1      | Not applicable | East US <br> East US 2 <br> North Central US <br> South Central US <br> West US <br> West US 3  | Not available       |
 Phi-4 <br>  Phi-4-mini-instruct <br>  Phi-4-multimodal-instruct    | Not applicable | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3  | East US 2 <br> East US <br> North Central US <br> South Central US <br> West US <br> West US 3      |
 Phi-3.5-vision-Instruct     | Not applicable | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3  | Not available       |
 Phi-3.5-MoE-Instruct     | Not applicable | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3  | East US 2 <br> East US <br> North Central US <br> South Central US <br> West US <br> West US 3      |

articles/ai-foundry/model-inference/concepts/models.md

@@ -141,10 +141,12 @@ See [this model collection in Azure AI Foundry portal](https://ai.azure.com/expl
 
 ### Microsoft
 
-Phi is a family of lightweight, state-of-the-art open models. These models were trained with Phi-3 datasets. The datasets include both synthetic data and the filtered, publicly available websites data, with a focus on high quality and reasoning-dense properties. The models underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
+Microsoft models include various model groups such as MAI models, Phi models, healthcare AI models, and more. To see all the available Microsoft models, view [the Microsoft model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
+
 
 | Model  | Type | Tier | Capabilities |
 | ------ | ---- | --- | ------------ |
+| [MAI-DS-R1](https://ai.azure.com/explore/models/MAI-DS-R1/version/1/registry/azureml) |  chat-completion <br /> [(with reasoning content)](../how-to/use-chat-reasoning.md) | Global standard | - **Input:** text (163,840 tokens) <br /> - **Output:**  (163,840 tokens) <br /> - **Languages:** `en` and `zh` <br />  - **Tool calling:** No <br /> - **Response formats:** Text. |
 | [Phi-4-mini-instruct](https://ai.azure.com/explore/models/Phi-4-mini-instruct/version/1/registry/azureml) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** `ar`, `zh`, `cs`, `da`, `nl`, `en`, `fi`, `fr`, `de`, `he`, `hu`, `it`, `ja`, `ko`, `no`, `pl`, `pt`, `ru`, `es`, `sv`, `th`, `tr`, and `uk` <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 | [Phi-4-multimodal-instruct](https://ai.azure.com/explore/models/Phi-4-multimodal-instruct/version/1/registry/azureml) | chat-completion | Global standard | - **Input:** text, images, and audio (131,072 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** `ar`, `zh`, `cs`, `da`, `nl`, `en`, `fi`, `fr`, `de`, `he`, `hu`, `it`, `ja`, `ko`, `no`, `pl`, `pt`, `ru`, `es`, `sv`, `th`, `tr`, and `uk` <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 | [Phi-4](https://ai.azure.com/explore/models/Phi-4/version/2/registry/azureml) | chat-completion | Global standard | - **Input:** text (16,384 tokens) <br /> - **Output:**  (16,384 tokens) <br /> - **Languages:** `en`, `ar`, `bn`, `cs`, `da`, `de`, `el`, `es`, `fa`, `fi`, `fr`, `gu`, `ha`, `he`, `hi`, `hu`, `id`, `it`, `ja`, `jv`, `kn`, `ko`, `ml`, `mr`, `nl`, `no`, `or`, `pa`, `pl`, `ps`, `pt`, `ro`, `ru`, `sv`, `sw`, `ta`, `te`, `th`, `tl`, `tr`, `uk`, `ur`, `vi`, `yo`, and `zh` <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
@@ -159,7 +161,8 @@ Phi is a family of lightweight, state-of-the-art open models. These models were
 | [Phi-3-small-128k-instruct](https://ai.azure.com/explore/models/Phi-3-small-128k-instruct/version/4/registry/azureml) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** `en` <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 
 
-See [this model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
+See [the Microsoft model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=phi).
+
 
 ### Mistral AI
 

articles/ai-services/openai/audio-completions-quickstart.md

@@ -45,14 +45,18 @@ recommendations: false
 
 ::: zone-end
 
-
 ## Clean-up resources
 
 If you want to clean up and remove an Azure OpenAI resource, you can delete the resource. Before deleting the resource, you must first delete any deployed models.
 
 - [Azure portal](../multi-service-resource.md?pivots=azportal#clean-up-resources)
 - [Azure CLI](../multi-service-resource.md?pivots=azcli#clean-up-resources)
 
+## Troubleshooting
+
+> [!NOTE]
+> When using `gpt-4o-audio-preview` for chat completions with the audio modality and `stream` is set to true the only supported audio format is pcm16.
+
 ## Related content
 
 * Learn more about Azure OpenAI [deployment types](./how-to/deployment-types.md).

articles/ai-services/openai/concepts/prompt-engineering.md

@@ -14,7 +14,7 @@ recommendations: false
 
 # Prompt engineering techniques
 
-GPT-3, GPT-3.5, GPT-4, and GPT-4o models from OpenAI are prompt-based. With prompt-based models, the user interacts with the model by entering a text prompt, to which the model responds with a text completion. This completion is the model’s continuation of the input text. These techniques are note recommended for o-series models.
+GPT-3, GPT-3.5, GPT-4, and GPT-4o models from OpenAI are prompt-based. With prompt-based models, the user interacts with the model by entering a text prompt, to which the model responds with a text completion. This completion is the model’s continuation of the input text. These techniques are not recommended for o-series models.
 
 While these models are extremely powerful, their behavior is also very sensitive to the prompt. This makes prompt construction an important skill to develop.
 

articles/ai-services/openai/includes/api-versions/latest-inference-preview.md

@@ -280,7 +280,7 @@ Creates a completion for the chat message
 | Name | Type | Description | Required | Default |
 |------|------|-------------|----------|---------|
 | audio | object | Parameters for audio output. Required when audio output is requested with<br>`modalities: ["audio"]`. <br> | No |  |
-| └─ format | enum | Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`,<br>`opus`, or `pcm16`. <br><br>Possible values: `wav`, `mp3`, `flac`, `opus`, `pcm16` | No |  |
+| └─ format | enum | Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`,<br>`opus`, or `pcm16`. <br><br>Possible values: `wav`, `mp3`, `flac`, `opus`, `pcm16`<br> When using `gpt-4o-audio-preview` and `stream` is set to true the only supported audio format is `pcm16`. | No |  |
 | └─ voice | enum | Specifies the voice type. Supported voices are `alloy`, `echo`, <br>`fable`, `onyx`, `nova`, and `shimmer`.<br><br>Possible values: `alloy`, `echo`, `fable`, `onyx`, `nova`, `shimmer` | No |  |
 | data_sources | array |   The configuration entries for Azure OpenAI chat extensions that use them.<br>  This additional specification is only compatible with Azure OpenAI. | No |  |
 | frequency_penalty | number | Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.<br> | No | 0 |