Dockerfile

# 官方纯 Linux-amd64 环境镜像
FROM ghcr.io/ggml-org/llama.cpp:server

WORKDIR /app

# 切换到 root 用户安装必要工具
USER root
RUN apt-get update && apt-get install -y --no-install-recommends \
    curl \
    ca-certificates \
    && rm -rf /var/lib/apt/lists/*

# 下载你的 Qwen 2.5 1.5B 优质轻量模型
RUN curl -L --fail \
    https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct-GGUF/resolve/main/qwen2.5-1.5b-instruct-q8_0.gguf \
    -o /app/model.gguf

# 暴露 Hugging Face 规定的 7860 端口
EXPOSE 7860

# 💡 核心修复点：删掉开头的 "/llama-server"，直接传参给底层的 ENTRYPOINT
CMD ["-m", "/app/model.gguf", \
     "--host", "0.0.0.0", \
     "--port", "7860", \
     "-ngl", "0", \
     "-t", "2", \
     "-c", "4096"]