[Hardware][TPU] Support MoE with Pallas GMM kernel (#6457)

This commit is contained in:
Woosuk Kwon
2024-07-16 09:56:28 -07:00
committed by GitHub
parent 9f4ccec761
commit c467dff24f
5 changed files with 89 additions and 8 deletions
+3 -1
View File
@@ -1,4 +1,4 @@
ARG NIGHTLY_DATE="20240601"
ARG NIGHTLY_DATE="20240713"
ARG BASE_IMAGE="us-central1-docker.pkg.dev/tpu-pytorch-releases/docker/xla:nightly_3.10_tpuvm_$NIGHTLY_DATE"
FROM $BASE_IMAGE
@@ -6,6 +6,8 @@ WORKDIR /workspace
# Install aiohttp separately to avoid build errors.
RUN pip install aiohttp
# Install NumPy 1 instead of NumPy 2.
RUN pip install "numpy<2"
# Install the TPU and Pallas dependencies.
RUN pip install torch_xla[tpu] -f https://storage.googleapis.com/libtpu-releases/index.html
RUN pip install torch_xla[pallas] -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html -f https://storage.googleapis.com/jax-releases/jaxlib_nightly_releases.html