feat(llama.cpp): add llama.cpp comprehensive recipes

Covers building, installing, running models with llama-server, and complete uninstallation of llama.cpp.

recipes/llama.cpp/01_install-llama-cpp.md

@@ -0,0 +1,77 @@
+# Install llama.cpp
+
+This recipe builds and installs llama.cpp from source on Debian/Ubuntu, including build dependencies and the `llama-server` binary.
+
+## Install Build Dependencies
+
+```bash
+sudo apt update
+sudo apt install -y build-essential cmake git libcurl4-openssl-dev
+```
+
+## Clone Repository
+
+```bash
+git clone https://github.com/ggml-org/llama.cpp
+cd llama.cpp
+```
+
+## Build Project
+
+```bash
+cmake -B build
+cmake --build build --config Release -j $(nproc)
+```
+
+## Install Binaries
+
+```bash
+sudo cmake --install build
+sudo ldconfig
+```
+
+## Verify Installation
+
+```bash
+llama-server --version
+```
+
+Expected output:
+
+```text
+version: xxxx (xxxxxxxx)
+```
+
+## Verify Binary Location
+
+```bash
+which llama-server
+```
+
+Expected output:
+
+```text
+/usr/local/bin/llama-server
+```
+
+## Troubleshooting
+
+### Command not found
+
+Verify that the binary was installed.
+
+```bash
+ls /usr/local/bin/llama-server
+```
+
+If it is missing, repeat the build and install steps.
+
+### Build fails with missing dependency
+
+Verify the build dependencies are installed.
+
+```bash
+dpkg -l | grep -E "build-essential|cmake|libcurl4-openssl-dev"
+```
+
+If any are missing, repeat the dependency installation step.

recipes/llama.cpp/02_run-models-on-llama-cpp.md

@@ -0,0 +1,94 @@
+# Run Models on llama.cpp
+
+This recipe runs a model with `llama-server` inside a persistent `tmux` session, downloading it directly from Hugging Face.
+
+## Install tmux
+
+```bash
+sudo apt install -y tmux
+```
+
+## Start a tmux Session
+
+```bash
+tmux new -s llama
+```
+
+## Run the Model Server
+
+Replace:
+
+- `<REPO>` with the Hugging Face model repo (e.g. `deepreinforce-ai/Ornith-1.0-9B-GGUF`)
+- `<HOST>` by default `0.0.0.0` hear all network interfaces
+- `<PORT>` with the port to serve on (e.g. `8000`)
+- `<CONTEXT_SIZE>` with the context window size (e.g. `262144`)
+
+```bash
+llama-server -hf <REPO> --host <HOST> --port <PORT> -c <CONTEXT_SIZE>
+```
+
+## Detach from the Session
+
+Press:
+
+```text
+Ctrl+B, D
+```
+
+## List tmux Sessions
+
+```bash
+tmux ls
+```
+
+Expected output:
+
+```text
+llama: 1 windows (created ...)
+```
+
+## Reattach to the Session
+
+```bash
+tmux attach -t llama
+```
+
+## Verify Server is Running
+
+Replace `<PORT>` with the configured port.
+
+```bash
+curl http://localhost:<PORT>/health
+```
+
+Expected output:
+
+```json
+{"status":"ok"}
+```
+
+## Troubleshooting
+
+### Port already in use
+
+Verify that no other process is bound to the port.
+
+```bash
+sudo ss -tulpn | grep <PORT>
+```
+
+### Session not found
+
+List active sessions and verify the name.
+
+```bash
+tmux ls
+```
+
+### Server not responding
+
+Reattach to the session and check the server logs.
+
+```bash
+tmux attach -t llama
+```

recipes/llama.cpp/03_uninstall-llama-cpp.md

@@ -0,0 +1,55 @@
+# Uninstall llama.cpp
+
+This recipe completely removes llama.cpp from Ubuntu, including installed binaries, libraries, and the cloned source repository.
+
+## Stop Running Sessions
+
+```bash
+tmux kill-session -t llama
+```
+
+## Remove Installed Binaries
+
+```bash
+sudo rm -f /usr/local/bin/llama*
+```
+
+## Remove Installed Libraries
+
+```bash
+sudo rm -f /usr/local/lib/libllama*
+sudo rm -f /usr/local/lib/libggml*
+sudo rm -rf /usr/local/include/llama.h /usr/local/include/ggml*
+sudo ldconfig
+```
+
+## Remove Cloned Repository
+
+Replace `<PATH>` with the path where the repository was cloned (e.g. `~/llama.cpp`).
+
+```bash
+rm -rf <PATH>
+```
+
+## Remove Build Dependencies (Optional)
+
+Only run this if the packages are not needed by other projects.
+
+```bash
+sudo apt purge -y build-essential cmake libcurl4-openssl-dev
+sudo apt autoremove -y
+```
+
+## Verify Removal
+
+Check that the binary is no longer available.
+
+```bash
+which llama
+```
+
+Expected output:
+
+```text
+(no output)
+```