Update Dockerfile

Dockerfile

@@ -4,44 +4,25 @@ FROM ubuntu:22.04
 RUN apt-get update && \
     apt-get install -y \
     build-essential \
-    libssl-dev \
-    zlib1g-dev \
-    libboost-all-dev \
-    libopenblas-dev \
-    libomp-dev \
+    python3-pip \
     cmake \
-    pkg-config \
     git \
-    python3-pip \
-    curl \
-    wget && \
+    curl && \
     rm -rf /var/lib/apt/lists/*
-    
-# Install Python dependencies with transfer acceleration
-RUN pip3 install huggingface-hub[hf_transfer] openai gradio duckduckgo-search
-
-# Build llama.cpp
-RUN git clone https://github.com/ggerganov/llama.cpp && \
-    cd llama.cpp && \
-    cmake -B build -S . -DLLAMA_BUILD_SERVER=ON -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS -DCMAKE_BUILD_TYPE=Release && \
-    cmake --build build --config Release --target llama-server -j $(nproc)
 
-# Download from official Qwen GGUF repo
-# We target the Q8_0 file for high precision
-RUN mkdir -p /models && \
-    HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download \
-    Qwen/Qwen3-0.6B-GGUF \
-    --include "*q8_0.gguf" \
-    --local-dir /models \
-    --local-dir-use-symlinks False
+# Set environment variables for faster downloads
+ENV HF_HUB_ENABLE_HF_TRANSFER=1
 
-# Rename the file to a standard name for our startup script
-RUN find /models -name "*.gguf" -exec mv {} /models/model.gguf \;
+# Install Python dependencies
+# llama-cpp-python will be compiled during install
+RUN pip3 install --upgrade pip && \
+    pip3 install llama-cpp-python huggingface-hub[hf_transfer] gradio duckduckgo-search
 
+# Copy the application code
 COPY app.py /app.py
-COPY start.sh /start.sh
-RUN chmod +x /start.sh
 
-EXPOSE 7860 8080
+# Expose the Gradio port
+EXPOSE 7860
 
-CMD ["/start.sh"]
+# Run the script directly
+CMD ["python3", "/app.py"]