@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / eeffde1ac0
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 18:15 UTC

[TPU] Upgrade PyTorch XLA nightly (#7967)

Woosuk Kwon committed Aug 28, 2024 at 13:10 UTC eeffde1ac01f575196655ad1cc8480b86967330b
2 files changed +5 -4
Dockerfile.tpu
+1 -1
@@ -1,4 +1,4 @@
1 -ARG NIGHTLY_DATE="20240808"
1 +ARG NIGHTLY_DATE="20240828"
2 ARG BASE_IMAGE="us-central1-docker.pkg.dev/tpu-pytorch-releases/docker/xla:nightly_3.10_tpuvm_$NIGHTLY_DATE"
3
4 FROM $BASE_IMAGE
docs/source/getting_started/tpu-installation.rst
+4 -3
@@ -56,9 +56,10 @@ First, install the dependencies:
56 $ pip uninstall torch torch-xla -y
57
58 $ # Install PyTorch and PyTorch XLA.
59 - $ export DATE="+20240808"
60 - $ pip install https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch-nightly${DATE}-cp310-cp310-linux_x86_64.whl
61 - $ pip install https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch_xla-nightly${DATE}-cp310-cp310-linux_x86_64.whl
59 + $ export DATE="20240828"
60 + $ export TORCH_VERSION="2.5.0"
61 + $ pip install https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch-${TORCH_VERSION}.dev${DATE}-cp310-cp310-linux_x86_64.whl
62 + $ pip3 install https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch_xla-${TORCH_VERSION}.dev${DATE}-cp310-cp310-linux_x86_64.whl
63
64 $ # Install JAX and Pallas.
65 $ pip install torch_xla[tpu] -f https://storage.googleapis.com/libtpu-releases/index.html