Merge pull request #2679 from santiagxf/santiagxf/deepseek-inference

feat: deepseek + reasoning models

Author: Stacyrch140

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

@@ -17,9 +17,6 @@ ms.custom: references_regions, tool_generated
 
 Azure AI model inference in Azure AI Foundry gives you access to flagship models in Azure AI to consume them as APIs without hosting them on your infrastructure.
 
-> [!TIP]
-> DeepSeek-R1 (preview) is available for deployment as [Serverless API endpoint](../../../ai-studio/how-to/deploy-models-deepseek.md).
-
 :::image type="content" source="../media/models/models-catalog.gif" alt-text="An animation showing Azure AI studio model catalog section and the models available." lightbox="../media/models/models-catalog.gif":::
 
 Model availability varies by model provider, deployment SKU, and cloud. All models available in Azure AI Model Inference support the [Global standard](deployment-types.md#global-standard) deployment type which uses global capacity to guarantee throughput. [Azure OpenAI models](#azure-openai) also support regional deployments and [sovereign clouds](/entra/identity-platform/authentication-national-cloud)—Azure Government, Azure Germany, and Azure China 21Vianet.
@@ -52,10 +49,11 @@ Azure OpenAI Service offers a diverse set of models with different capabilities
 - Models that can transcribe and translate speech to text
 
 | Model  | Type | Tier | Capabilities |
-| ------ | ---- | --- | ------------ |
+| ------ | ---- | ---- | ------------ |
+| [o3-mini](https://ai.azure.com/explore/models/o3-mini/version/2025-01-31/registry/azure-openai) | chat-completion | Global standard | - **Input:** text and image (200,000 tokens) <br /> - **Output:** text (100,000 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
 | [o1](https://ai.azure.com/explore/models/o1/version/2024-12-17/registry/azure-openai) | chat-completion | Global standard | - **Input:** text and image (200,000 tokens) <br /> - **Output:** text (100,000 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
 | [o1-preview](https://ai.azure.com/explore/models/o1-preview/version/1/registry/azure-openai) | chat-completion | Global standard <br />Standard<br /> | - **Input:** text (128,000 tokens) <br /> - **Output:**  (32,768 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
-| [o1-mini](https://ai.azure.com/explore/models/o1-mini/version/1/registry/azure-openai) | chat-completion | Global standard <br />Standard | - **Input:** text (128,000 tokens) <br /> - **Output:**  (65,536 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
+| [o1-mini](https://ai.azure.com/explore/models/o1-mini/version/1/registry/azure-openai) | chat-completion | Global standard <br />Standard | - **Input:** text (128,000 tokens) <br /> - **Output:**  (65,536 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 | [gpt-4o-realtime-preview](https://ai.azure.com/explore/models/gpt-4o-realtime-preview/version/2024-10-01/registry/azure-openai) | real-time | Global standard | - **Input:** control, text, and audio (131,072 tokens) <br /> - **Output:** text and audio (16,384 tokens) <br /> - **Languages:** en <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [gpt-4o](https://ai.azure.com/explore/models/gpt-4o/version/2024-11-20/registry/azure-openai) | chat-completion | Global standard <br />Standard<br />Batch<br />Provisioned<br />Global provisioned<br />Data Zone | - **Input:** text and image (131,072 tokens) <br /> - **Output:** text (16,384 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
 | [gpt-4o-mini](https://ai.azure.com/explore/models/gpt-4o-mini/version/2024-07-18/registry/azure-openai) | chat-completion | Global standard <br />Standard<br />Batch<br />Provisioned<br />Global provisioned<br />Data Zone | - **Input:** text, image, and audio (131,072 tokens) <br /> - **Output:**  (16,384 tokens) <br /> - **Languages:** `en`, `it`, `af`, `es`, `de`, `fr`, `id`, `ru`, `pl`, `uk`, `el`, `lv`, `zh`, `ar`, `tr`, `ja`, `sw`, `cy`, `ko`, `is`, `bn`, `ur`, `ne`, `th`, `pa`, `mr`, and `te`. <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON, structured outputs |
@@ -93,6 +91,16 @@ Core42 includes autoregressive bi-lingual LLMs for Arabic & English with state-o
 
 See [this model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=core42).
 
+### DeepSeek
+
+DeepSeek family of models include DeepSeek-R1, which excels at reasoning tasks using a step-by-step training process, such as language, scientific reasoning, and coding tasks.
+
+| Model  | Type | Tier | Capabilities |
+| ------ | ---- | --- | ------------ |
+| [DeekSeek-R1](https://ai.azure.com/explore/models/deepseek-r1/version/1/registry/azureml-deepseek) | chat-completion | Global standard | - **Input:** text (16,384 tokens) <br /> - **Output:**  (163,840 tokens) <br /> - **Languages:** `en` and `zh` <br />  - **Tool calling:** No <br /> - **Response formats:** Text (with reasoning content). |
+
+See [this model collection in Azure AI Foundry portal](https://ai.azure.com/explore/models?&selectedCollection=deepseek).
+
 ### Meta
 
 Meta Llama models and tools are a collection of pretrained and fine-tuned generative AI text and image reasoning models. Meta models range is scale to include:
@@ -143,10 +151,10 @@ Mistral AI offers two categories of models: premium models including Mistral Lar
 | Model  | Type | Tier | Capabilities |
 | ------ | ---- | --- | ------------ |
 | [Ministral-3B](https://ai.azure.com/explore/models/Ministral-3B/version/1/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Languages:** fr, de, es, it, and en <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
-| [Mistral-large](https://ai.azure.com/explore/models/Mistral-large/version/1/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (32,768 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** fr, de, es, it, and en <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
+| [Mistral-large](https://ai.azure.com/explore/models/Mistral-large/version/1/registry/azureml-mistral) <br /> (deprecated) | chat-completion | Global standard | - **Input:** text (32,768 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** fr, de, es, it, and en <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Mistral-small](https://ai.azure.com/explore/models/Mistral-small/version/1/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (32,768 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Languages:** fr, de, es, it, and en <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Mistral-Nemo](https://ai.azure.com/explore/models/Mistral-Nemo/version/1/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Languages:** en, fr, de, es, it, zh, ja, ko, pt, nl, and pl <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
-| [Mistral-large-2407](https://ai.azure.com/explore/models/Mistral-large-2407/version/1/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** en, fr, de, es, it, zh, ja, ko, pt, nl, and pl <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
+| [Mistral-large-2407](https://ai.azure.com/explore/models/Mistral-large-2407/version/1/registry/azureml-mistral) <br /> (legacy) | chat-completion | Global standard | - **Input:** text (131,072 tokens) <br /> - **Output:**  (4,096 tokens) <br /> - **Languages:** en, fr, de, es, it, zh, ja, ko, pt, nl, and pl <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Mistral-Large-2411](https://ai.azure.com/explore/models/Mistral-Large-2411/version/2/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (128,000 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Languages:** en, fr, de, es, it, zh, ja, ko, pt, nl, and pl <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Codestral-2501](https://ai.azure.com/explore/models/Codestral-2501/version/2/registry/azureml-mistral) | chat-completion | Global standard | - **Input:** text (262,144 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Languages:** en <br />  - **Tool calling:** No <br /> - **Response formats:** Text |
 

articles/ai-foundry/model-inference/how-to/use-chat-reasoning.md

@@ -0,0 +1,53 @@
+---
+title: How to use reasoning models with Azure AI model inference
+titleSuffix: Azure AI Foundry
+description: Learn how to use reasoning capabilities from models with Azure AI model inference
+manager: scottpolly
+author: msakande
+reviewer: santiagxf
+ms.service: azure-ai-model-inference
+ms.topic: how-to
+ms.date: 1/31/2025
+ms.author: mopeakande
+ms.reviewer: fasantia
+ms.custom: generated
+zone_pivot_groups: azure-ai-inference-samples
+---
+
+# How to use reasoning models with Azure AI model inference
+
+
+::: zone pivot="programming-language-python"
+
+[!INCLUDE [python](../includes/use-chat-reasoning/python.md)]
+::: zone-end
+
+
+::: zone pivot="programming-language-javascript"
+
+[!INCLUDE [javascript](../includes/use-chat-reasoning/javascript.md)]
+::: zone-end
+
+
+::: zone pivot="programming-language-java"
+
+[!INCLUDE [java](../includes/use-chat-reasoning/java.md)]
+::: zone-end
+
+
+::: zone pivot="programming-language-csharp"
+
+[!INCLUDE [csharp](../includes/use-chat-reasoning/csharp.md)]
+::: zone-end
+
+
+::: zone pivot="programming-language-rest"
+
+[!INCLUDE [rest](../includes/use-chat-reasoning/rest.md)]
+::: zone-end
+
+## Related content
+
+* [Use embeddings models](use-embeddings.md)
+* [Use image embeddings models](use-image-embeddings.md)
+* [Azure AI Model Inference API](.././reference/reference-model-inference-api.md)

articles/ai-foundry/model-inference/includes/use-chat-completions/csharp.md

@@ -26,15 +26,12 @@ To use chat completion models in your application, you need:
 
 * A chat completions model deployment. If you don't have one read [Add and configure models to Azure AI services](../../how-to/create-model-deployments.md) to add a chat completions model to your resource.
 
-* Install the Azure AI inference package with the following command:
+* Install the [Azure AI inference package](https://aka.ms/azsdk/azure-ai-inference/python/reference) with the following command:
 
     ```bash
     dotnet add package Azure.AI.Inference --prerelease
     ```
 
-    > [!TIP]
-    > Read more about the [Azure AI inference package and reference](https://aka.ms/azsdk/azure-ai-inference/python/reference).
-
 * If you are using Entra ID, you also need the following package:
 
     ```bash

articles/ai-foundry/model-inference/includes/use-chat-completions/java.md

@@ -26,7 +26,7 @@ To use chat completion models in your application, you need:
 
 * A chat completions model deployment. If you don't have one read [Add and configure models to Azure AI services](../../how-to/create-model-deployments.md) to add a chat completions model to your resource.
 
-* Add the Azure AI inference package to your project:
+* Add the [Azure AI inference package](https://aka.ms/azsdk/azure-ai-inference/java/reference) to your project:
 
   ```xml
   <dependency>
@@ -36,9 +36,6 @@ To use chat completion models in your application, you need:
   </dependency>
   ```
 
-  > [!TIP]
-  > Read more about the [Azure AI inference package and reference](https://aka.ms/azsdk/azure-ai-inference/java/reference).
-
 * If you are using Entra ID, you also need the following package:
 
   ```xml

articles/ai-foundry/model-inference/includes/use-chat-completions/javascript.md

@@ -26,15 +26,12 @@ To use chat completion models in your application, you need:
 
 * A chat completions model deployment. If you don't have one read [Add and configure models to Azure AI services](../../how-to/create-model-deployments.md) to add a chat completions model to your resource.
 
-* Install the Azure Inference library for JavaScript with the following command:
+* Install the [Azure Inference library for JavaScript](https://aka.ms/azsdk/azure-ai-inference/javascript/reference) with the following command:
 
   ```bash
   npm install @azure-rest/ai-inference
   ```
       
-  > [!TIP]
-  > Read more about the [Azure AI inference package and reference](https://aka.ms/azsdk/azure-ai-inference/javascript/reference).
-
 ## Use chat completions
 
 First, create the client to consume the model. The following code uses an endpoint URL and key that are stored in environment variables.

articles/ai-foundry/model-inference/includes/use-chat-completions/python.md

@@ -26,15 +26,12 @@ To use chat completion models in your application, you need:
 
 * A chat completions model deployment. If you don't have one read [Add and configure models to Azure AI services](../../how-to/create-model-deployments.md) to add a chat completions model to your resource.
 
-* Install the Azure AI inference package with the following command:
+* Install the [Azure AI inference package for Python](https://aka.ms/azsdk/azure-ai-inference/python/reference) with the following command:
 
   ```bash
   pip install -U azure-ai-inference
   ```
 
-  > [!TIP]
-  > Read more about the [Azure AI inference package and reference](https://aka.ms/azsdk/azure-ai-inference/python/reference).
-
 ## Use chat completions
 
 First, create the client to consume the model. The following code uses an endpoint URL and key that are stored in environment variables.