Revise GPU Dockerfile

Dockerfile.gpu

@@ -1,35 +1,32 @@
-FROM nvidia/cuda:12.5.1-cudnn-runtime-ubuntu24.04
+FROM nvidia/cuda:12.6.3-cudnn-runtime-ubuntu24.04
+
+# Install 'uv', which we will use to install Python dependencies
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
 ENV PYTHONDONTWRITEBYTECODE="1" \
     PYTHONUNBUFFERED="1" \
     DEBIAN_FRONTEND="noninteractive"
 
-# Install system dependencies
-RUN apt-get update && \
-    apt-get install -y --no-install-recommends \
-        python3-pip \
-        python-is-python3 && \
-    apt-get clean && \
-    rm -rf /var/lib/apt/lists/*
-
-# hadolint ignore=DL3013
-RUN python3 -m pip install --no-cache-dir --upgrade pip setuptools wheel
-
+# Install Python and create venv
 WORKDIR /app/
+RUN uv python install 3.12
+RUN uv venv
+RUN uv pip install --no-cache-dir --upgrade pip setuptools wheel
 
 # Download models during build instead of copying from local
 COPY scripts/model_download.bash /tmp/model_download.bash
-RUN python3 -m pip install --no-cache-dir huggingface-hub && \
+RUN . .venv/bin/activate && \
+    uv pip install --no-cache-dir huggingface-hub && \
     bash /tmp/model_download.bash && \
     rm /tmp/model_download.bash
 
 # Install CPU requirements
 COPY requirements.cpu.txt ./
-RUN python3 -m pip install --no-cache-dir -r ./requirements.cpu.txt
+RUN uv pip install --no-cache-dir -r ./requirements.cpu.txt
 
 # Install GPU PyTorch requirements
 COPY requirements.torch.gpu.txt ./
-RUN python3 -m pip install --no-cache-dir -r ./requirements.torch.gpu.txt
+RUN uv pip install --no-cache-dir -r ./requirements.torch.gpu.txt
 
 COPY app ./app
 COPY main.py ./