Qwency / Dockerfile
Promise Emmanuel Oluwadare
Fix HF build: add compiler toolchain for llama-cpp-python
ec1e88d
Raw
History Blame Contribute Delete
923 Bytes
FROM python:3.10-slim
ENV PYTHONDONTWRITEBYTECODE=1 \
PYTHONUNBUFFERED=1 \
PIP_NO_CACHE_DIR=1 \
PIP_DISABLE_PIP_VERSION_CHECK=1 \
PIP_DEFAULT_TIMEOUT=120 \
HF_HOME=/tmp/hf-cache \
HF_HUB_ENABLE_HF_TRANSFER=1
RUN apt-get update && apt-get install -y --no-install-recommends \
build-essential \
ca-certificates \
cmake \
curl \
git \
libgomp1 \
ninja-build \
pkg-config \
&& rm -rf /var/lib/apt/lists/*
WORKDIR /app
COPY requirements.txt /app/requirements.txt
RUN python -m pip install --upgrade pip setuptools wheel && \
pip install --prefer-binary -r /app/requirements.txt
COPY start_server.py /app/start_server.py
ENV PORT=7860 \
MODEL_REPO=unsloth/Qwen3.5-0.8B-GGUF \
MODEL_FILE=Qwen3.5-0.8B-Q4_K_M.gguf \
MODEL_DIR=/tmp/models \
N_CTX=2048 \
N_THREADS=2 \
CHAT_FORMAT=chatml
EXPOSE 7860
CMD ["python", "/app/start_server.py"]