Update Dockerfile

Dockerfile

@@ -1,39 +1,39 @@
-# Use official Python 3.11 base image
-FROM python:3.11
+# Use official Python image with basic system utilities
+FROM python:3.11-slim
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y \
+    wget \
     build-essential \
-    cmake \
-    git \
-    curl \
     && rm -rf /var/lib/apt/lists/*
 
 # Set working directory
 WORKDIR /app
 
-# Download the quantized GGUF model from Hugging Face
-RUN curl -L -o /app/tinyllama_dop_q4_k_m.gguf \
-    https://huggingface.co/Kalpokoch/FinetunedQuantizedTinyLama/resolve/main/tinyllama_dop_q4_k_m.gguf
-
-# Set Hugging Face and ChromaDB cache directories
+# Set HF cache dir for transformers, chromadb, etc.
 ENV TRANSFORMERS_CACHE=/app/.cache \
     HF_HOME=/app/.cache \
-    PYTHONUNBUFFERED=1 \
-    PERSIST_DIR=/app/vector_database
+    PIP_NO_CACHE_DIR=true
+
+# Create required dirs with open permissions
+RUN mkdir -p /app/.cache /app/vector_database && chmod -R 777 /app/.cache /app/vector_database
 
-RUN mkdir -p /app/.cache && chmod -R 777 /app/.cache
-RUN mkdir -p ${PERSIST_DIR} && chmod -R 777 ${PERSIST_DIR}
+# Pre-download your quantized GGUF model from HF
+# (Replace with your exact file path if different)
+RUN wget https://huggingface.co/Kalpokoch/FinetunedQuantizedTinyLama/resolve/main/tinyllama_dop_q4_k_m.gguf -O /app/tinyllama_dop_q4_k_m.gguf
 
-# Install Python dependencies
+# Install Python dependencies separately to leverage Docker caching
 COPY requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
+RUN pip install -r requirements.txt
+
+# Install llama-cpp-python from wheel (avoid compilation)
+RUN pip install llama-cpp-python --prefer-binary
 
-# Copy project code
+# Copy the app code
 COPY . .
 
-# Expose the port FastAPI will run on
+# Expose port for FastAPI
 EXPOSE 7860
 
-# Run the FastAPI app using uvicorn
+# Launch app
 CMD ["uvicorn", "app.app:app", "--host", "0.0.0.0", "--port", "7860"]