@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 3ed5e73146
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, 05:45 UTC

[TPU] Update requirements-tpu (#10726)

Signed-off-by: Richard Liu <ricliu@google.com>

Richard Liu committed Nov 28, 2024 at 02:30 UTC 3ed5e7314667f0a9c0c47e6d635ac82fd93296a2
1 file changed +5 -5
requirements-tpu.txt
+5 -5
@@ -16,8 +16,8 @@ ray[default]
16 --find-links https://storage.googleapis.com/libtpu-releases/index.html
17 --find-links https://storage.googleapis.com/jax-releases/jax_nightly_releases.html
18 --find-links https://storage.googleapis.com/jax-releases/jaxlib_nightly_releases.html
19 -torch==2.6.0.dev20241114+cpu
20 -torchvision==0.20.0.dev20241114+cpu
21 -torch_xla[tpu] @ https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch_xla-2.6.0.dev20241114-cp310-cp310-linux_x86_64.whl
22 -jaxlib==0.4.32.dev20240829
23 -jax==0.4.32.dev20240829
19 +torch==2.6.0.dev20241126+cpu
20 +torchvision==0.20.0.dev20241126+cpu
21 +torch_xla[tpu] @ https://storage.googleapis.com/pytorch-xla-releases/wheels/tpuvm/torch_xla-2.6.0.dev20241126-cp310-cp310-linux_x86_64.whl
22 +jaxlib==0.4.36.dev20241122
23 +jax==0.4.36.dev20241122