Update Dockerfile

Dockerfile

@@ -1,21 +1,32 @@
 FROM python:3.12-slim
 
-# Install system dependencies
-RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
+# Avoid interactive prompts during apt install
+ENV DEBIAN_FRONTEND=noninteractive
 
-# Install CPU-only PyTorch + vLLM
+# Use a more stable Debian mirror and retry on failures
+RUN sed -i 's|deb.debian.org|deb.debian.net|g' /etc/apt/sources.list && \
+    apt-get update -o Acquire::Retries=3 && \
+    apt-get install -y --no-install-recommends git && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install CPU-only PyTorch and vLLM
 RUN pip install --no-cache-dir torch==2.4.0 --index-url https://download.pytorch.org/whl/cpu
 RUN pip install --no-cache-dir vllm
 
-# Expose port
+# Expose desired port
 EXPOSE 7860
 
-# Env variables
+# Environment variables for host/port
 ENV VLLM_HOST=0.0.0.0
 ENV VLLM_PORT=7860
-ENV HUGGING_FACE_HUB_TOKEN=<your_hf_token>
 
-# Command to run vLLM on CPU
-CMD ["sh", "-c", "vllm serve --model unsloth/llama-2-7b-bnb-4bit --device cpu --host $VLLM_HOST --port $VLLM_PORT"]
+# Hugging Face token for private or gated models
+ENV HUGGING_FACE_HUB_TOKEN=<your_hf_token>
 
+# Optional: store HF cache in RAM-only volume
+ENV HF_HOME=/tmp/.cache/huggingface
+RUN mkdir -p /tmp/.cache/huggingface && chmod -R 777 /tmp/.cache/huggingface
+VOLUME ["/tmp/.cache/huggingface"]
 
+# Command: serve the model on CPU
+CMD ["sh", "-c", "vllm serve --model unsloth/llama-2-7b-bnb-4bit --device cpu --host $VLLM_HOST --port $VLLM_PORT"]