Revert and add Cmake

Dockerfile

@@ -18,22 +18,18 @@ RUN useradd -m -u 1000 user
 # Set the working directory
 WORKDIR /app
 
-# Create and activate virtual environment
-RUN python3 -m venv /app/venv
-ENV PATH="/app/venv/bin:$PATH"
-
 # Copy the requirements file
 COPY requirements.txt .
 
 # Install Python dependencies
-RUN . /app/venv/bin/activate && \
-    pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir -r requirements.txt
+RUN pip3 install --no-cache-dir --upgrade pip && \
+    pip3 install --no-cache-dir -r requirements.txt
 
 # Install llama-cpp-python with CUDA support
-RUN . /app/venv/bin/activate && \
-    pip install llama-cpp-python \
-    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu121
+ENV FORCE_CMAKE=1
+ENV CMAKE_ARGS="-DLLAMA_CUBLAS=on"
+RUN pip3 install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu121
 
 # Copy the rest of the application code
 COPY . .
@@ -48,4 +44,4 @@ USER user
 ENV PATH="/home/user/.local/bin:$PATH"
 
 # Run the application
-CMD ["/app/venv/bin/uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]