diff --git a/.github/workflows/release-docker-npu-nightly.yml b/.github/workflows/release-docker-npu-nightly.yml index 5855fd729..0089f1d98 100644 --- a/.github/workflows/release-docker-npu-nightly.yml +++ b/.github/workflows/release-docker-npu-nightly.yml @@ -80,6 +80,6 @@ jobs: push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }} provenance: false build-args: | - SGLANG_KERNEL_NPU_TAG=2026.9.0 + SGLANG_KERNEL_NPU_TAG=2026.9.0.post4 CANN_VERSION=${{ matrix.cann_version }} DEVICE_TYPE=${{ matrix.device_type }} diff --git a/.github/workflows/release-docker-npu.yml b/.github/workflows/release-docker-npu.yml index eaa13da93..0e8573d78 100644 --- a/.github/workflows/release-docker-npu.yml +++ b/.github/workflows/release-docker-npu.yml @@ -87,7 +87,7 @@ jobs: push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }} provenance: false build-args: | - SGLANG_KERNEL_NPU_TAG=2026.9.0 + SGLANG_KERNEL_NPU_TAG=2026.9.0.post4 CANN_VERSION=${{ matrix.cann_version }} DEVICE_TYPE=${{ matrix.device_type }} SGLANG_TAG=${{ steps.version.outputs.version }} diff --git a/scripts/ci/npu/npu_ci_install_dependency.sh b/scripts/ci/npu/npu_ci_install_dependency.sh index 192530517..2186837b1 100755 --- a/scripts/ci/npu/npu_ci_install_dependency.sh +++ b/scripts/ci/npu/npu_ci_install_dependency.sh @@ -55,7 +55,7 @@ ${PIP_INSTALL} triton-ascend==3.2.1.dev20260530 --extra-index-url=https://mirror ### Install sgl-kernel-npu -SGLANG_KERNEL_NPU_TAG="2026.9.0" +SGLANG_KERNEL_NPU_TAG="2026.9.0.post4" mkdir sgl-kernel-npu (cd sgl-kernel-npu && wget "${GITHUB_PROXY_URL:=""}https://github.com/sgl-project/sgl-kernel-npu/releases/download/${SGLANG_KERNEL_NPU_TAG}/sgl-kernel-npu-${SGLANG_KERNEL_NPU_TAG}-torch${PYTORCH_VERSION}-py311-cann9.0.0-${DEVICE_TYPE}-$(arch).zip" \ && unzip ./sgl-kernel-npu-${SGLANG_KERNEL_NPU_TAG}-torch${PYTORCH_VERSION}-py311-cann9.0.0-${DEVICE_TYPE}-$(arch).zip \