Create Dockerfile

Dockerfile

@@ -0,0 +1,35 @@
+FROM ghcr.io/ggml-org/llama.cpp:full
+
+WORKDIR /app
+
+RUN apt update && apt install -y python3 python3-pip python3-venv curl
+
+RUN python3 -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+RUN pip install -U pip huggingface_hub gradio requests
+
+# Download SmolLM2-1.7B-Instruct GGUF (Q4_K_M ~1GB, runs well on free CPU)
+RUN python3 -c '\
+from huggingface_hub import hf_hub_download; \
+hf_hub_download(\
+  repo_id="bartowski/SmolLM2-1.7B-Instruct-GGUF", \
+  filename="SmolLM2-1.7B-Instruct-Q4_K_M.gguf", \
+  local_dir="/app"\
+)'
+
+COPY app.py /app/app.py
+
+# Start llama.cpp server in background, then launch Gradio
+CMD bash -c '\
+  /llama-server \
+    -m /app/SmolLM2-1.7B-Instruct-Q4_K_M.gguf \
+    --host 0.0.0.0 \
+    --port 8080 \
+    -t 2 \
+    -c 4096 \
+    -n 2048 \
+    --log-disable &
+  sleep 10 && \
+  python3 /app/app.py\
+'