diff --git a/Dockerfile b/Dockerfile
index 9b65dd5..8d9b8ea 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -8,3 +8,6 @@ COPY utils.py /usr/local/lib/python3.12/dist-packages/vllm/tool_parsers/utils.py
 # Patch hf renderer to force string content format for GLM models
 # This fixes the issue where tool response content is dropped
 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