add Compose how-to page for Docker Model Runner support with Compose

Signed-off-by: Guillaume Lours <[email protected]>

Author: glours

content/manuals/compose/how-tos/model-runner.md

@@ -0,0 +1,66 @@
+---
+title: Using Docker Model Runner with Compose
+description: Learn how to integrate Docker Model Runner with Docker Compose to build AI-powered applications
+keywords: compose, docker compose, model runner, ai, llm, artificial intelligence, machine learning
+weight: 20
+---
+
+## Using Docker Model Runner with Compose
+
+Docker Model Runner can be integrated with Docker Compose to run AI models as part of your multi-container applications.  
+This allows you to define and run AI-powered applications alongside your other services.
+
+### Prerequisites
+
+- Docker Compose v2.35 or later
+- Docker Desktop 4.41 or later 
+- Docker Model Runner enabled in Docker Desktop
+- Apple Silicon Mac (currently Model Runner is only available for Mac with Apple Silicon)
+
+### Enabling Docker Model Runner
+
+Before you can use Docker Model Runner with Compose, you need to enable it in Docker Desktop, as described in the [Docker Model Runner documentation](/manuals/desktop/features/model-runner/).
+
+### Provider services
+
+Compose introduces a new service type called `provider` that allows you to declare platform capabilities required by your application. For AI models, you can use the `model` type to declare model dependencies.
+
+Here's an example of how to define a model provider:
+
+```yaml
+services:
+  chat:
+    image: my-chat-app
+    depends_on:
+      - ai-runner
+
+  ai-runner:
+    provider:
+      type: model
+      options:
+        model: ai/smollm2
+```
+
+You should notice the dedicated `provider` attribute in the `ai-runner` service.   
+This attribute specifies that the service is a model provider and let you define options such as the name of the model to be used.
+
+There is also a `depends_on` attribute in the `chat` service.  
+This attribute specifies that the `chat` service depends on the `ai-runner` service.  
+This means that the `ai-runner` service will be started before the `chat` service to allow injection of model information to the `chat` service.
+
+### How it works
+
+During the `docker compose up` process, Docker Model Runner will automatically pull and run the specified model.  
+It will also send to Compose the model tag name and the URL to access the model runner.
+
+Those information will be then pass to services which declare a dependency on the model provider.  
+In the example above, the `chat` service will receive 2 env variables prefixed by the service name:
+ - `AI-RUNNER_URL` with the URL to access the model runner
+ - `AI-RUNNER_MODEL` with the model name which could be passed with the URL to request the model.
+
+This allows the `chat` service to interact with the model and use it for its own purposes.
+
+
+For more information, see:
+- [Docker Model Runner documentation](/manuals/desktop/features/model-runner/)
+