diff --git a/Dockerfile b/Dockerfile index 4f6f11c..b56db9f 100644 --- a/Dockerfile +++ b/Dockerfile @@ -11,7 +11,11 @@ COPY vllm_patches/hf.py /usr/local/lib/python3.12/dist-packages/vllm/renderers/h # Install LMCache for KV cache offloading / sharing across nodes # Build from source to match CUDA 13 (pip wheel is CUDA 12) -RUN git clone --depth 1 https://github.com/LMCache/LMCache.git /tmp/lmcache && \ +RUN apt-get update && apt-get install -y --no-install-recommends git && \ + git clone --depth 1 https://github.com/LMCache/LMCache.git /tmp/lmcache && \ cd /tmp/lmcache && \ pip install --no-cache-dir --no-build-isolation -e . && \ - rm -rf /tmp/lmcache + rm -rf /tmp/lmcache && \ + apt-get remove -y git && \ + apt-get autoremove -y && \ + rm -rf /var/lib/apt/lists/*