Update README.md

README.md

@@ -1,3 +1,34 @@
 # Alpine LLaMA
 
-Alpine LLaMA is an ultra-compact Docker image (less than 10 MB), providing a LLaMA.cpp HTTP server to run languages models inference. It is particularly suitable for environments with limited disk space and no GPUs.
+This is an ultra-compact Docker image (less than 10 MB), providing a [LLaMA.cpp](https://github.com/ggerganov/llama.cpp) HTTP server for language model inference.
+
+## When to use it?
+
+This Docker image is particularly suited for:
+- Alpine-based application images, which require a HTTP inference server.
+- Servers that cannot do GPU-accelerated inference, e.g. a CPU-only VPS.
+- Environments with limited disk space.
+
+## Examples of use
+
+### Standalone, online model
+
+You can start a local standalone HTTP inference server who listens at the port 50000, leveraging a Qwen2.5 Coder 1.5B quantized model hosted, for example on [Hugging Face](https://huggingface.co/):
+
+```bash
+docker run --name alpine-llama -p 50000:8080 -e LLAMA_ARG_MODEL_URL=https://huggingface.co/bartowski/Qwen2.5-Coder-1.5B-Instruct-GGUF/resolve/main/Qwen2.5-Coder-1.5B-Instruct-Q4_K_M.gguf -e LLAMA_ARG_CTX_SIZE=2048 samueltallet/alpine-llama-cpp-server
+```
+
+Once the GGUF model file is downloaded (and cached in the Docker container), you can query the OpenAI-compatible Chat Completions API endpoint exposed, say using cURL:
+
+```bash
+curl http://127.0.0.1:50000/v1/chat/completions \
+  -d '{
+    "messages": [
+      { "role": "user", "content": "Explain GraphDB with simple words." }
+    ],
+    "temperature": 0.3,
+    "max_tokens": 100
+  }'
+# > GraphDB is a type of database that stores and manages data in a way that is similar to the way people think about things in the world. It uses a graph data model, which means that it stores data in nodes and edges, where nodes represent entities and edges represent relationships between those entities [...]
+```