feat: update Dockerfile (#1033)

Co-authored-by: vhain <vhain6512@gmail.com>

.github/workflows/release-docker.yml

-name: Release Docker
+name: Release Docker Images
 on:
   push:
     branches:
@@ -14,25 +14,28 @@ jobs:
     environment: 'prod'
     strategy:
       matrix:
-        cuda_version: ['12.1.1', '12.4.1']
+        cuda_version: ['11.8.0', '12.1.1', '12.4.1']
+        build_type: ['all', 'srt']
     steps:
       - name: Delete huge unnecessary tools folder
         run: rm -rf /opt/hostedtoolcache
 
       - name: Checkout repository
         uses: actions/checkout@v3
-      
+
       - name: Login to Docker Hub
         uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
-      
+
       - name: Build and Push
         run: |
           version=$(cat python/sglang/version.py | cut -d'"' -f2)
-          
-          if [ "${{ matrix.cuda_version }}" = "12.1.1" ]; then
+
+          if [ "${{ matrix.cuda_version }}" = "11.8.0" ]; then
+            cuda_tag="cu118"
+          elif [ "${{ matrix.cuda_version }}" = "12.1.1" ]; then
             cuda_tag="cu121"
           elif [ "${{ matrix.cuda_version }}" = "12.4.1" ]; then
             cuda_tag="cu124"
@@ -40,13 +43,22 @@ jobs:
             echo "Unsupported CUDA version"
             exit 1
           fi
-          
+
           tag=v${version}-${cuda_tag}
-          
-          docker build . -f docker/Dockerfile --build-arg CUDA_VERSION=${{ matrix.cuda_version }} -t lmsysorg/sglang:${tag} --no-cache
-          docker push lmsysorg/sglang:${tag}
-          
+
+          if [ "${{ matrix.build_type }}" = "all" ]; then
+            tag_suffix=""
+          elif [ "${{ matrix.build_type }}" = "srt" ]; then
+            tag_suffix="-srt"
+          else
+            echo "Unsupported build type"
+            exit 1
+          fi
+
+          docker build . -f docker/Dockerfile --build-arg CUDA_VERSION=${{ matrix.cuda_version }} --build-arg BUILD_TYPE=${{ matrix.build_type }} -t lmsysorg/sglang:${tag}${tag_suffix} --no-cache
+          docker push lmsysorg/sglang:${tag}${tag_suffix}
+
           if [ "${{ matrix.cuda_version }}" = "12.1.1" ]; then
-            docker tag lmsysorg/sglang:${tag} lmsysorg/sglang:latest
-            docker push lmsysorg/sglang:latest
+            docker tag lmsysorg/sglang:${tag}${tag_suffix} lmsysorg/sglang:latest${tag_suffix}
+            docker push lmsysorg/sglang:latest${tag_suffix}
           fi

docker/Dockerfile

 ARG CUDA_VERSION=12.1.1
-
-FROM nvidia/cuda:${CUDA_VERSION}-devel-ubuntu20.04
-
-ARG PYTHON_VERSION=3
-
+FROM nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu20.04
+ARG BUILD_TYPE=all
 ENV DEBIAN_FRONTEND=noninteractive
 
 RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
     && echo 'tzdata tzdata/Zones/America select Los_Angeles' | debconf-set-selections \
-    && apt-get update -y \
-    && apt-get install -y ccache software-properties-common \
-    && add-apt-repository ppa:deadsnakes/ppa \
-    && apt-get update -y \
-    && apt-get install -y python${PYTHON_VERSION} python${PYTHON_VERSION}-dev python${PYTHON_VERSION}-venv python${PYTHON_VERSION}-pip \
-    && if [ "${PYTHON_VERSION}" != "3" ]; then update-alternatives --install /usr/bin/python3 python3 /usr/bin/python${PYTHON_VERSION} 1; fi \
+    && apt update -y \
+    && apt install software-properties-common -y \
+    && add-apt-repository ppa:deadsnakes/ppa -y && apt update \
+    && apt install python3.10 -y \
+    && update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.8 1 && update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.10 2 \
+    && update-alternatives --set python3 /usr/bin/python3.10 && apt install python3.10-distutils -y \
+    && apt install curl git sudo -y \
+    && curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && python3 get-pip.py \
     && python3 --version \
     && python3 -m pip --version \
     && rm -rf /var/lib/apt/lists/* \
-    && apt-get clean
-
-RUN apt-get update -y \
-    && apt-get install -y git curl sudo
+    && apt clean
 
 WORKDIR /sgl-workspace
 
-RUN pip3 --no-cache-dir install --upgrade pip \
-    && pip3 --no-cache-dir install --upgrade setuptools wheel \
+RUN python3 -m pip install --upgrade pip setuptools wheel html5lib six \
     && git clone --depth=1 https://github.com/sgl-project/sglang.git \
     && cd sglang \
-    && pip --no-cache-dir install -e "python[all]" \
-    && pip3 --no-cache-dir install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/
+    && if [ "$BUILD_TYPE" = "srt" ]; then \
+         python3 -m pip --no-cache-dir install -e "python[srt]"; \
+       else \
+         python3 -m pip --no-cache-dir install -e "python[all]"; \
+       fi
+
+ARG CUDA_VERSION
+RUN if [ "$CUDA_VERSION" = "12.1.1" ]; then \
+        export CUDA_IDENTIFIER=cu121 && \
+        python3 -m pip --no-cache-dir install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/; \
+    elif [ "$CUDA_VERSION" = "12.4.1" ]; then \
+        export CUDA_IDENTIFIER=cu124 && \
+        python3 -m pip --no-cache-dir install flashinfer -i https://flashinfer.ai/whl/cu124/torch2.4/; \
+    elif [ "$CUDA_VERSION" = "11.8.0" ]; then \
+        export CUDA_IDENTIFIER=cu118 && \
+        python3 -m pip install torch==2.4.0 --index-url https://download.pytorch.org/whl/cu118 && \
+        python3 -m pip --no-cache-dir install flashinfer -i https://flashinfer.ai/whl/cu118/torch2.4/; \
+    else \
+        echo "Unsupported CUDA version: $CUDA_VERSION" && exit 1; \
+    fi
+
+RUN python3 -m pip cache purge
 
 ENV DEBIAN_FRONTEND=interactive