support 12.5 CUDA runtime (#3231)

.github/workflows/release-docker.yml

@@ -14,7 +14,7 @@ jobs:
     environment: 'prod'
     strategy:
       matrix:
-        cuda_version: ['11.8.0', '12.1.1', '12.4.1']
+        cuda_version: ['11.8.0', '12.1.1', '12.4.1', '12.5.1']
         build_type: ['all', 'srt']
     steps:
       - name: Delete huge unnecessary tools folder
@@ -39,6 +39,8 @@ jobs:
             cuda_tag="cu121"
           elif [ "${{ matrix.cuda_version }}" = "12.4.1" ]; then
             cuda_tag="cu124"
+          elif [ "${{ matrix.cuda_version }}" = "12.5.1" ]; then
+            cuda_tag="cu125"
           else
             echo "Unsupported CUDA version"
             exit 1
@@ -58,7 +60,7 @@ jobs:
           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 [ "${{ matrix.cuda_version }}" = "12.4.1" ]; then
+          if [ "${{ matrix.cuda_version }}" = "12.5.1" ]; then
             docker tag lmsysorg/sglang:${tag}${tag_suffix} lmsysorg/sglang:latest${tag_suffix}
             docker push lmsysorg/sglang:latest${tag_suffix}
           fi

docker/Dockerfile

@@ -30,6 +30,8 @@ RUN python3 -m pip install --upgrade pip setuptools wheel html5lib six \
          python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu121; \
        elif [ "$CUDA_VERSION" = "12.4.1" ]; then \
          python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu124; \
+       elif [ "$CUDA_VERSION" = "12.5.1" ]; then \
+         python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu124; \
        elif [ "$CUDA_VERSION" = "11.8.0" ]; then \
          python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu118; \
          python3 -m pip install sgl-kernel -i https://docs.sglang.ai/whl/cu118; \
@@ -42,6 +44,8 @@ RUN python3 -m pip install --upgrade pip setuptools wheel html5lib six \
            python3 -m pip --no-cache-dir install -e "python[srt]" --find-links https://flashinfer.ai/whl/cu121/torch2.4/flashinfer/; \
          elif [ "$CUDA_VERSION" = "12.4.1" ]; then \
            python3 -m pip --no-cache-dir install -e "python[srt]" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer/; \
+         elif [ "$CUDA_VERSION" = "12.5.1" ]; then \
+           python3 -m pip --no-cache-dir install -e "python[srt]" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer/; \
          elif [ "$CUDA_VERSION" = "11.8.0" ]; then \
            python3 -m pip --no-cache-dir install -e "python[srt]" --find-links https://flashinfer.ai/whl/cu118/torch2.4/flashinfer/; \
            python3 -m pip install sgl-kernel -i https://docs.sglang.ai/whl/cu118; \
@@ -53,6 +57,8 @@ RUN python3 -m pip install --upgrade pip setuptools wheel html5lib six \
            python3 -m pip --no-cache-dir install -e "python[all]" --find-links https://flashinfer.ai/whl/cu121/torch2.4/flashinfer/; \
          elif [ "$CUDA_VERSION" = "12.4.1" ]; then \
            python3 -m pip --no-cache-dir install -e "python[all]" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer/; \
+         elif [ "$CUDA_VERSION" = "12.5.1" ]; then \
+           python3 -m pip --no-cache-dir install -e "python[all]" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer/; \
          elif [ "$CUDA_VERSION" = "11.8.0" ]; then \
            python3 -m pip --no-cache-dir install -e "python[all]" --find-links https://flashinfer.ai/whl/cu118/torch2.4/flashinfer/; \
            python3 -m pip install sgl-kernel -i https://docs.sglang.ai/whl/cu118; \