Dockerfile

# Read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
# Dockerfile for Qwen-VL PicExam API with CPU inference optimization

FROM python:3.10-slim

# 安装系统依赖
RUN apt-get update && apt-get install -y \
    git \
    wget \
    curl \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# 创建用户
RUN useradd -m -u 1000 user
USER user
ENV PATH="/home/user/.local/bin:$PATH"

# 设置工作目录
WORKDIR /app

# 设置环境变量优化内存使用
ENV PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:512
ENV TOKENIZERS_PARALLELISM=false
ENV OMP_NUM_THREADS=4
ENV MKL_NUM_THREADS=4

# 复制并安装 Python 依赖
COPY --chown=user ./requirements.txt requirements.txt
RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir --upgrade -r requirements.txt

# 复制应用代码
COPY --chown=user . /app

# 暴露端口
EXPOSE 7860

# 启动命令，增加内存和超时配置
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860", "--timeout-keep-alive", "300", "--workers", "1"]