Use Flashinfer 0.6.18 release for CUDA 13.4 package (#37073)
This commit is contained in:
@@ -99,7 +99,7 @@ ARG PIP_DEFAULT_INDEX
|
|||||||
ARG UBUNTU_MIRROR
|
ARG UBUNTU_MIRROR
|
||||||
ARG GITHUB_ARTIFACTORY=github.com
|
ARG GITHUB_ARTIFACTORY=github.com
|
||||||
ARG INSTALL_FLASHINFER_JIT_CACHE=1
|
ARG INSTALL_FLASHINFER_JIT_CACHE=1
|
||||||
ARG FLASHINFER_VERSION=0.6.18rc10
|
ARG FLASHINFER_VERSION=0.6.18
|
||||||
ARG CUTLASS_DSL_VERSION=4.8.0.dev0
|
ARG CUTLASS_DSL_VERSION=4.8.0.dev0
|
||||||
ARG MOONCAKE_VERSION=0.3.12.post1
|
ARG MOONCAKE_VERSION=0.3.12.post1
|
||||||
ARG MSCCLPP_VERSION=sglang-v0.9.1
|
ARG MSCCLPP_VERSION=sglang-v0.9.1
|
||||||
@@ -894,6 +894,10 @@ elif [ "${CUDA_VERSION%%.*}" = "13" ]; then \
|
|||||||
python3 -m pip install "cuda-python>=13,<14" ; \
|
python3 -m pip install "cuda-python>=13,<14" ; \
|
||||||
fi
|
fi
|
||||||
|
|
||||||
|
# Install the latest available AI Dynamo prerelease from NVIDIA's package index.
|
||||||
|
RUN --mount=type=cache,target=/root/.cache/pip \
|
||||||
|
python3 -m pip install --pre --extra-index-url https://pypi.nvidia.com/ ai-dynamo
|
||||||
|
|
||||||
# Add yank script
|
# Add yank script
|
||||||
COPY --chown=root:root --chmod=755 docker/configs/yank /usr/local/bin/yank
|
COPY --chown=root:root --chmod=755 docker/configs/yank /usr/local/bin/yank
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user