biondizzle/vllm-glm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix lmcache

Browse Source
This commit is contained in:
biondizzle
2026-04-15 07:25:23 +00:00
parent 0b70c975bd
commit 64784741de
1 changed files with 5 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
Dockerfile
View File

@@ -10,4 +10,8 @@ COPY utils.py /usr/local/lib/python3.12/dist-packages/vllm/tool_parsers/utils.py
COPY vllm_patches/hf.py /usr/local/lib/python3.12/dist-packages/vllm/renderers/hf.py
# Install LMCache for KV cache offloading / sharing across nodes
RUN pip install --no-cache-dir lmcache
# 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 && \
cd /tmp/lmcache && \
pip install --no-cache-dir --no-build-isolation -e . && \
rm -rf /tmp/lmcache