Lemonade Playbook for Local AI on CPU, GPU, and NPU

playbooks/dependencies/lemonade.md

@@ -2,41 +2,68 @@
 
 #### Installing Lemonade
 
-Download and install Lemonade Server from [lemonade-server.ai](https://lemonade-server.ai).
+<!-- @os:windows -->
+Download the latest installer from [lemonade-server.ai](https://github.com/lemonade-sdk/lemonade/releases/latest/download/lemonade.msi) and run the `.msi` file. The installer adds `lemonade-server` to your system PATH automatically.
 
-#### Starting Lemonade
+You can also install silently from the command line:
 
-<!-- @os:windows -->
-<!-- @test:id=lemonade-chat-gpt-oss-120b timeout=1200 hidden=True -->
-```powershell
-lemonade-server --version
-winget upgrade -e --id AMD.LemonadeServer
-lemonade-server --version
+```cmd
+msiexec /i lemonade-server-minimal.msi /qn
 ```
-<!-- @test:end -->
 <!-- @os:end -->
 
-<!-- @os:windows -->
-
-1. Open PowerShell or Command Prompt
-2. Start Lemonade with gpt-oss-120b:
-```cmd
-lemonade-server run gpt-oss-120b-mxfp4-GGUF
+<!-- @os:linux -->
+**Ubuntu (snap):**
+```bash
+sudo snap install lemonade-server
 ```
 
-The server starts on `http://localhost:8000` with an OpenAI-compatible API at `/api/v1`.
+**Arch Linux (AUR):**
+```bash
+yay -S lemonade-server
+```
 
+For other distributions or to install from source, see the [full installation options](https://lemonade-server.ai/install_options.html).
 <!-- @os:end -->
 
-<!-- @os:linux -->
+#### Verifying Lemonade Installation
 
-Start Lemonade with gpt-oss-120b and ROCm backend:
+Open a terminal and run:
+
+```
+lemonade-server --version
+```
+
+You should see output like:
+
+```
+lemonade-server x.y.z
+```
+
+If you see a version number, Lemonade is installed and ready to go.
+
+
+#### Starting Lemonade
+
+To start the server, open a terminal and run:
 ```bash
-lemonade-server run gpt-oss-120b-mxfp4-GGUF --llamacpp rocm
+lemonade-server serve
 ```
 
 The server starts on `http://localhost:8000` with an OpenAI-compatible API at `/api/v1`.
 
+To run a specific model immediately, use the `run` command:
+
+```bash
+lemonade-server run Gemma-3-4b-it-GGUF
+```
+
 > **Tip**: Use `lemonade-server list` to see available models, or `lemonade-server pull <MODEL_NAME>` to download new ones.
 
-<!-- @os:end -->
+To start Lemonade server with ROCm backend:
+
+```bash
+lemonade-server serve --llamacpp rocm
+```
+
+For the latest installation options or troubleshooting, please refer to the official Lemonade documentation.