diff --git a/.devops/llama-server-cuda.Dockerfile b/.devops/llama-server-cuda.Dockerfile
index bf8a198f99f73..cb9c17edbed44 100644
--- a/.devops/llama-server-cuda.Dockerfile
+++ b/.devops/llama-server-cuda.Dockerfile
@@ -1,20 +1,27 @@
-ARG UBUNTU_VERSION=22.04
+ARG UBUNTU_VERSION=20.04
+
 # This needs to generally match the container host's environment.
-ARG CUDA_VERSION=12.6.0
+ARG CUDA_VERSION=11.4.0
 # Target the CUDA build image
-ARG BASE_CUDA_DEV_CONTAINER=nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
+ARG BASE_CUDA_DEV_CONTAINER=registry.cn-hangzhou.aliyuncs.com/reg_pub/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
 # Target the CUDA runtime image
-ARG BASE_CUDA_RUN_CONTAINER=nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}
+ARG BASE_CUDA_RUN_CONTAINER=registry.cn-hangzhou.aliyuncs.com/reg_pub/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}
 
 FROM ${BASE_CUDA_DEV_CONTAINER} AS build
-
+ARG DEBIAN_FRONTEND=noninteractive
+ENV TZ=Africa/Johannesburg 
 # CUDA architecture to build for (defaults to all supported archs)
 ARG CUDA_DOCKER_ARCH=default
+ADD https://cmake.org/files/v3.25/cmake-3.25.2-linux-x86_64.tar.gz /tmp/
+WORKDIR /tmp
+RUN tar -xf cmake-3.25.2-linux-x86_64.tar.gz  && \
+     cp -rd cmake-3.25.2-linux-x86_64/bin /usr/local/ && \
+     cp -rd cmake-3.25.2-linux-x86_64/doc /usr/local/  && \
+     cp -rd cmake-3.25.2-linux-x86_64/share /usr/local/ && \
+     rm -r /tmp/*
+RUN  apt-get update && \
+    apt-get install -y build-essential git  libcurl4-openssl-dev && \
 
-RUN apt-get update && \
-    apt-get install -y build-essential git cmake libcurl4-openssl-dev
-
-WORKDIR /app
 
 COPY . .