File size: 1,040 Bytes
e68d948 aa48f13 e68d948 aa48f13 e68d948 1171a9d a882caf 9b2fbac e68d948 a882caf 9b2fbac e68d948 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | FROM ghcr.io/ggml-org/llama.cpp:full
WORKDIR /app
# Python仮想環境を作るための最低限のものを入れる
RUN apt-get update && apt-get install -y --no-install-recommends \
python3-pip python3-venv \
&& rm -rf /var/lib/apt/lists/*
# venv を作成
RUN python3 -m venv /opt/venv
ENV PATH="/opt/venv/bin:$PATH"
# venv にだけ pip パッケージを入れる
RUN pip install --no-cache-dir -U huggingface_hub
# モデルとmmprojをダウンロード
RUN python3 -c 'from huggingface_hub import hf_hub_download; \
repo="unsloth/Qwen3.6-27B-GGUF"; \
hf_hub_download(repo_id=repo, filename="Qwen3.6-27B-IQ4_NL.gguf", local_dir="/app"); \
hf_hub_download(repo_id=repo, filename="mmproj-BF16.gguf", local_dir="/app")'
CMD ["--server", \
"-m", "/app/Qwen3.6-27B-IQ4_NL.gguf", \
"--mmproj", "/app/mmproj-BF16.gguf", \
"--host", "0.0.0.0", \
"--port", "7860", \
"-t", "2", \
"--cache-type-k", "q8_0", \
"--cache-type-v", "iq4_nl", \
"-c", "64000", \
"-n", "38912"] |