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