chore: cleanup docker image (#10671)

.github/workflows/release-docker-dev.yml

@@ -12,15 +12,9 @@ jobs:
     strategy:
       matrix:
         variant:
-          - version: 12.6.1
+          - version: 12.9.1
             type: all
             tag: dev
-          - version: 12.8.1
-            type: blackwell
-            tag: blackwell
-          - version: 12.9.1
-            type: blackwell
-            tag: b200-cu129
 
     steps:
       - name: Checkout repository

.github/workflows/release-docker.yml

@@ -14,15 +14,8 @@ jobs:
     environment: 'prod'
     strategy:
       matrix:
-        cuda_version: ['12.6.1', '12.8.1', '12.9.1']
+        cuda_version: ['12.6.1', '12.9.1']
-        build_type: ['all', 'blackwell']
+        build_type: ['all']
-        exclude:
-          - cuda_version: '12.6.1'
-            build_type: 'blackwell'
-          - cuda_version: '12.8.1'
-            build_type: 'all'
-          - cuda_version: '12.9.1'
-            build_type: 'all'
     steps:
       - name: Delete huge unnecessary tools folder
         run: rm -rf /opt/hostedtoolcache
@@ -84,7 +77,6 @@ jobs:
           fi
 
           docker buildx build --output type=image,compression=zstd . -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 [ "${{ matrix.cuda_version }}" = "12.6.1" ]; then
             docker tag lmsysorg/sglang:${tag}${tag_suffix} lmsysorg/sglang:latest${tag_suffix}

python/pyproject.toml

@@ -67,6 +67,7 @@ dependencies = [
   "tiktoken",
   "anthropic>=0.20.0",
   "torch_memory_saver==0.0.8",
+  "nvidia-cutlass-dsl==4.2.0",
 ]
 
 [project.optional-dependencies]
@@ -89,9 +90,9 @@ tracing = [
     "opentelemetry-exporter-otlp-proto-grpc",
 ]
 all = ["sglang[test]", "sglang[decord]"]
-blackwell = ["nvidia-cutlass-dsl==4.2.0", "sglang[test]", "sglang[decord]"]
+blackwell = ["sglang[test]", "sglang[decord]"]
-blackwell_aarch64 = ["nvidia-cutlass-dsl==4.2.0", "sglang[test]"]
+blackwell_aarch64 = ["sglang[test]"]
-dev = ["sglang[test]"]
+dev = ["sglang[test]", "sglang[decord]"]
 
 [project.urls]
 "Homepage" = "https://github.com/sgl-project/sglang"