diff --git a/.github/workflows/_schedule_image_build.yaml b/.github/workflows/_schedule_image_build.yaml
index 108bbf26666..76bbe6f6ca2 100644
--- a/.github/workflows/_schedule_image_build.yaml
+++ b/.github/workflows/_schedule_image_build.yaml
@@ -76,6 +76,13 @@ jobs:
         driver: docker-container
         use: true
 
+    - name: Configure sccache
+      uses: actions/github-script@v7
+      with:
+        script: |
+          core.exportVariable('ACTIONS_RESULTS_URL', process.env.ACTIONS_RESULTS_URL || '');
+          core.exportVariable('ACTIONS_RUNTIME_TOKEN', process.env.ACTIONS_RUNTIME_TOKEN || '');
+
     - name: Build and push
       uses: docker/build-push-action@v7
       id: build
@@ -89,6 +96,11 @@ jobs:
         outputs: type=image,name=quay.io/ascend/vllm-ascend,push-by-digest=true,name-canonical=true,push=${{ inputs.should_push }}
         build-args: |
           PIP_INDEX_URL=https://pypi.org/simple
+          SCCACHE_GHA_ENABLED=true
+          MATRIX_ARCH=${{ matrix.arch }}
+        secrets: |
+          ACTIONS_RESULTS_URL=${{ env.ACTIONS_RESULTS_URL }}
+          ACTIONS_RUNTIME_TOKEN=${{ env.ACTIONS_RUNTIME_TOKEN }}
         provenance: false
 
     - name: Export digest
diff --git a/Dockerfile b/Dockerfile
index 6132ffa5816..457c1a6a75f 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -20,6 +20,8 @@ FROM quay.io/ascend/cann:8.5.1-910b-ubuntu22.04-py3.11
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG MOONCAKE_TAG="v0.3.8.post1"
 ARG SOC_VERSION="ascend910b1"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 # Define environments
 ENV DEBIAN_FRONTEND=noninteractive
@@ -31,17 +33,28 @@ WORKDIR /workspace
 
 COPY . /vllm-workspace/vllm-ascend/
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 # Install Mooncake dependencies
-RUN apt-get update -y && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    apt-get update -y && \
     apt-get install -y git vim wget net-tools gcc g++ cmake libnuma-dev libjemalloc2 && \
     git clone --depth 1 --branch ${MOONCAKE_TAG} https://github.com/kvcache-ai/Mooncake /vllm-workspace/Mooncake && \
     cp /vllm-workspace/vllm-ascend/tools/mooncake_installer.sh /vllm-workspace/Mooncake/ && \
     cd /vllm-workspace/Mooncake && bash mooncake_installer.sh -y && \
     ARCH=$(uname -m) && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/devlib:/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/lib64:$LD_LIBRARY_PATH && \
-    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON && \
+    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON -DENABLE_SCCACHE=ON && \
     make -j$(nproc) && make install && \
+    sccache --show-stats && \
     rm -fr /vllm-workspace/Mooncake/build && \
     rm -rf /var/cache/apt/* && \
     rm -rf /var/lib/apt/lists/*
@@ -60,11 +73,19 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -v -e /vllm-workspace/vllm
 # Install vllm-ascend
 # Append `libascend_hal.so` path (devlib) to LD_LIBRARY_PATH
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    sccache --show-stats && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/Dockerfile.310p b/Dockerfile.310p
index 023397977ca..5f93d676edc 100644
--- a/Dockerfile.310p
+++ b/Dockerfile.310p
@@ -19,6 +19,8 @@ FROM quay.io/ascend/cann:8.5.1-310p-ubuntu22.04-py3.11
 
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG SOC_VERSION="ascend310p1"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 # Define environments
 ENV DEBIAN_FRONTEND=noninteractive
@@ -36,6 +38,10 @@ WORKDIR /workspace
 
 COPY . /vllm-workspace/vllm-ascend/
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 RUN pip config set global.index-url ${PIP_INDEX_URL}
 
 # Install vLLM
@@ -50,11 +56,19 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -v -e /vllm-workspace/vllm
 # Install vllm-ascend
 # Append `libascend_hal.so` path (devlib) to LD_LIBRARY_PATH
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    sccache --show-stats && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/Dockerfile.310p.openEuler b/Dockerfile.310p.openEuler
index de5594ce2bf..7154b18dbc6 100644
--- a/Dockerfile.310p.openEuler
+++ b/Dockerfile.310p.openEuler
@@ -19,6 +19,8 @@ FROM quay.io/ascend/cann:8.5.1-310p-openeuler24.03-py3.11
 
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG SOC_VERSION="ascend310p1"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 ENV SOC_VERSION=$SOC_VERSION \
     TASK_QUEUE_ENABLE=1 \
@@ -34,6 +36,10 @@ WORKDIR /workspace
 
 COPY . /vllm-workspace/vllm-ascend/
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 # Install vLLM
 ARG VLLM_REPO=https://github.com/vllm-project/vllm.git
 ARG VLLM_TAG=v0.18.0
@@ -45,12 +51,20 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -e /vllm-workspace/vllm/[a
 
 # Install vllm-ascend
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/c++/12:/usr/include/c++/12/`uname -i`-openEuler-linux && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    sccache --show-stats && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/Dockerfile.a3 b/Dockerfile.a3
index 45fe4f1e06b..f69290fcf5f 100644
--- a/Dockerfile.a3
+++ b/Dockerfile.a3
@@ -20,6 +20,8 @@ FROM quay.io/ascend/cann:8.5.1-a3-ubuntu22.04-py3.11
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG MOONCAKE_TAG=v0.3.8.post1
 ARG SOC_VERSION="ascend910_9391"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 COPY . /vllm-workspace/vllm-ascend/
 # Define environments
@@ -32,17 +34,28 @@ RUN pip config set global.index-url ${PIP_INDEX_URL}
 
 WORKDIR /workspace
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 # Install Mooncake dependencies
-RUN apt-get update -y && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    apt-get update -y && \
     apt-get install -y git vim wget net-tools gcc g++ cmake libnuma-dev libjemalloc2 && \
     git clone --depth 1 --branch ${MOONCAKE_TAG} https://github.com/kvcache-ai/Mooncake /vllm-workspace/Mooncake && \
     cp /vllm-workspace/vllm-ascend/tools/mooncake_installer.sh /vllm-workspace/Mooncake/ && \
     cd /vllm-workspace/Mooncake && bash mooncake_installer.sh -y && \
     ARCH=$(uname -m) && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/devlib:/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/lib64:$LD_LIBRARY_PATH && \
-    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON && \
+    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON -DENABLE_SCCACHE=ON && \
     make -j$(nproc) && make install && \
+    sccache --show-stats && \
     rm -fr /vllm-workspace/Mooncake/build && \
     rm -rf /var/cache/apt/* && \
     rm -rf /var/lib/apt/lists/*
@@ -59,11 +72,18 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -v -e /vllm-workspace/vllm
 # Install vllm-ascend
 # Append `libascend_hal.so` path (devlib) to LD_LIBRARY_PATH
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/Dockerfile.a3.openEuler b/Dockerfile.a3.openEuler
index fe30bdcc688..308edd8e717 100644
--- a/Dockerfile.a3.openEuler
+++ b/Dockerfile.a3.openEuler
@@ -20,6 +20,8 @@ FROM quay.io/ascend/cann:8.5.1-a3-openeuler24.03-py3.11
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG MOONCAKE_TAG="v0.3.8.post1"
 ARG SOC_VERSION="ascend910_9391"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 ENV SOC_VERSION=$SOC_VERSION \
     TASK_QUEUE_ENABLE=1 \
@@ -31,20 +33,31 @@ WORKDIR /workspace
 
 COPY . /vllm-workspace/vllm-ascend/
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 SHELL ["/bin/bash", "-c"]
 
-RUN yum update -y && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    yum update -y && \
     yum install -y git vim wget net-tools gcc gcc-c++ make cmake numactl-devel jemalloc && \
     git clone --depth 1 --branch ${MOONCAKE_TAG} https://github.com/kvcache-ai/Mooncake /vllm-workspace/Mooncake && \
     cp /vllm-workspace/vllm-ascend/tools/mooncake_installer.sh /vllm-workspace/Mooncake/ && \
     ARCH=$(uname -m) && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/devlib:/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/lib64:$LD_LIBRARY_PATH && \
     export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/c++/12:/usr/include/c++/12/${ARCH}-openEuler-linux && \
     cd /vllm-workspace/Mooncake && \
     bash mooncake_installer.sh -y && \
-    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON && \
+    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON -DENABLE_SCCACHE=ON && \
     make -j$(nproc) && make install && \
+    sccache --show-stats && \
     rm -fr /vllm-workspace/Mooncake/build && \
     rm -rf /var/cache/yum/*
 
@@ -59,12 +72,20 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -e /vllm-workspace/vllm/[a
 
 # Install vllm-ascend
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/c++/12:/usr/include/c++/12/`uname -i`-openEuler-linux && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    sccache --show-stats && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/Dockerfile.openEuler b/Dockerfile.openEuler
index c9608e1966f..f229024bca0 100644
--- a/Dockerfile.openEuler
+++ b/Dockerfile.openEuler
@@ -20,6 +20,8 @@ FROM quay.io/ascend/cann:8.5.1-910b-openeuler24.03-py3.11
 ARG PIP_INDEX_URL="https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 ARG MOONCAKE_TAG="v0.3.8.post1"
 ARG SOC_VERSION="ascend910b1"
+ARG SCCACHE_GHA_ENABLED
+ARG MATRIX_ARCH
 
 ENV SOC_VERSION=$SOC_VERSION \
     TASK_QUEUE_ENABLE=1 \
@@ -31,20 +33,31 @@ WORKDIR /workspace
 
 COPY . /vllm-workspace/vllm-ascend/
 
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    bash /vllm-workspace/vllm-ascend/tools/sccache_installer.sh
+
 SHELL ["/bin/bash", "-c"]
 
-RUN yum update -y && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    yum update -y && \
     yum install -y git vim wget net-tools gcc gcc-c++ make cmake numactl-devel jemalloc && \
     git clone --depth 1 --branch ${MOONCAKE_TAG} https://github.com/kvcache-ai/Mooncake /vllm-workspace/Mooncake && \
     cp /vllm-workspace/vllm-ascend/tools/mooncake_installer.sh /vllm-workspace/Mooncake/ && \
     ARCH=$(uname -m) && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/devlib:/usr/local/Ascend/ascend-toolkit/latest/${ARCH}-linux/lib64:$LD_LIBRARY_PATH && \
     export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/c++/12:/usr/include/c++/12/${ARCH}-openEuler-linux && \
     cd /vllm-workspace/Mooncake && \
     bash mooncake_installer.sh -y && \
-    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON && \
+    mkdir -p build && cd build && cmake .. -DUSE_ASCEND_DIRECT=ON -DENABLE_SCCACHE=ON && \
     make -j$(nproc) && make install && \
+    sccache --show-stats && \
     rm -fr /vllm-workspace/Mooncake/build && \
     rm -rf /var/cache/yum/*
 
@@ -59,12 +72,20 @@ RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install -e /vllm-workspace/vllm/[a
 
 # Install vllm-ascend
 # Installing vllm-ascend on x86 can pull upstream triton back in alongside triton-ascend. Remove it immediately after this step.
-RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+RUN --mount=type=secret,id=ACTIONS_RESULTS_URL,required=false \
+    --mount=type=secret,id=ACTIONS_RUNTIME_TOKEN,required=false \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
     source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
     source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "") && \
+    export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "") && \
+    export ACTIONS_CACHE_SERVICE_V2=on && \
+    export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED} && \
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/c++/12:/usr/include/c++/12/`uname -i`-openEuler-linux && \
     python3 -m pip install -v -e /vllm-workspace/vllm-ascend/ --extra-index https://download.pytorch.org/whl/cpu/ && \
+    sccache --show-stats && \
+    rm -f /usr/bin/sccache && \
     if [ "$(uname -i)" = "x86_64" ]; then python3 -m pip uninstall -y triton; fi && \
     python3 -m pip cache purge
 
diff --git a/csrc/build.sh b/csrc/build.sh
index 76efeaaa072..ed29c47dbda 100644
--- a/csrc/build.sh
+++ b/csrc/build.sh
@@ -178,7 +178,7 @@ CUSTOM_OPTION="${CUSTOM_OPTION} -DCUSTOM_ASCEND_CANN_PACKAGE_PATH=${ASCEND_CANN_
 set_env
 clean
 
-ccache_system=$(which ccache || true)
+ccache_system=$(command -v sccache 2>/dev/null || command -v ccache 2>/dev/null || true)
 if [ -n "${ccache_system}" ];then
     CUSTOM_OPTION="${CUSTOM_OPTION} -DENABLE_CCACHE=ON -DCUSTOM_CCACHE=${ccache_system}"
     gen_bisheng ${ccache_system}
diff --git a/csrc/cmake/config.cmake b/csrc/cmake/config.cmake
index b352a8bf850..1189d34933d 100644
--- a/csrc/cmake/config.cmake
+++ b/csrc/cmake/config.cmake
@@ -189,7 +189,7 @@ if (BUILD_OPEN_PROJECT)
         if (CUSTOM_CCACHE)
             set(CCACHE_PROGRAM ${CUSTOM_CCACHE})
         else()
-            find_program(CCACHE_PROGRAM ccache)
+            find_program(CCACHE_PROGRAM NAMES sccache ccache)
         endif ()
         if (CCACHE_PROGRAM)
             set(CMAKE_C_COMPILER_LAUNCHER   ${CCACHE_PROGRAM} CACHE PATH "C cache Compiler")
diff --git a/tools/sccache_installer.sh b/tools/sccache_installer.sh
new file mode 100644
index 00000000000..1dfe2ce68c6
--- /dev/null
+++ b/tools/sccache_installer.sh
@@ -0,0 +1,48 @@
+#!/usr/bin/env bash
+
+#
+# Copyright (c) 2026 Huawei Technologies Co., Ltd. All Rights Reserved.
+# Copyright 2023 The vLLM team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# This file is a part of the vllm-ascend project.
+
+set -e
+
+SCCACHE_VERSION="v0.14.0"
+
+echo "MATRIX_ARCH=${MATRIX_ARCH}"
+case "${MATRIX_ARCH}" in
+    linux/arm64) SCCACHE_ARCH="aarch64" ;;
+    linux/amd64) SCCACHE_ARCH="x86_64" ;;
+    *) echo "Unsupported TARGETPLATFORM for sccache: ${MATRIX_ARCH}" >&2; exit 1 ;;
+esac
+
+echo "SCCACHE_ARCH=${SCCACHE_ARCH}"
+SCCACHE_PACKAGE="sccache-${SCCACHE_VERSION}-${SCCACHE_ARCH}-unknown-linux-musl"
+SCCACHE_DOWNLOAD_URL="${SCCACHE_DOWNLOAD_URL:-https://github.com/mozilla/sccache/releases/download/${SCCACHE_VERSION}/${SCCACHE_PACKAGE}.tar.gz}"
+echo "SCCACHE_DOWNLOAD_URL=${SCCACHE_DOWNLOAD_URL}"
+
+curl -L -o /tmp/sccache.tar.gz "${SCCACHE_DOWNLOAD_URL}"
+tar -xzf /tmp/sccache.tar.gz -C /tmp
+mv "/tmp/${SCCACHE_PACKAGE}/sccache" /usr/bin/sccache
+chmod +x /usr/bin/sccache
+rm -rf /tmp/sccache.tar.gz "/tmp/${SCCACHE_PACKAGE}"
+
+export ACTIONS_RESULTS_URL=$(cat /run/secrets/ACTIONS_RESULTS_URL 2>/dev/null || echo "")
+export ACTIONS_RUNTIME_TOKEN=$(cat /run/secrets/ACTIONS_RUNTIME_TOKEN 2>/dev/null || echo "")
+export ACTIONS_CACHE_SERVICE_V2=on
+export SCCACHE_GHA_ENABLED=${SCCACHE_GHA_ENABLED}
+
+sccache --start-server
+sccache --show-stats