diff --git a/Dockerfile b/Dockerfile
index 0eb5a85..f344e1d 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -22,9 +22,9 @@ RUN --mount=type=cache,target=/root/.cache/pip \
 # Install torch and vllm based on CUDA version
 RUN if [[ "${WORKER_CUDA_VERSION}" == 11.8* ]]; then \
         python3.11 -m pip install -U --force-reinstall torch==2.1.2 xformers==0.0.23.post1 --index-url https://download.pytorch.org/whl/cu118; \
-        python3.11 -m pip install -e git+https://github.com/runpod/vllm-fork-for-sls-worker.git@cuda-11.8#egg=vllm; \
+        python3.11 -m pip install -e git+https://github.com/runpod/vllm-fork-for-sls-worker.git@old-11.8#egg=vllm; \
     else \
-        python3.11 -m pip install -e git+https://github.com/runpod/vllm-fork-for-sls-worker.git#egg=vllm; \
+        python3.11 -m pip install -e git+https://github.com/runpod/vllm-fork-for-sls-worker.git@old-12.1#egg=vllm; \
     fi && \
     rm -rf /root/.cache/pip