Owner

AI-Hypercomputer

已索引 14 個儲存庫 · 在 GitHub 檢視

• maxtext

  A simple, performant, and scalable Jax LLM!

  Python · 2421 個 star

• JetStream

  JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs wel...

  Python · 458 個 star

• maxdiffusion

  Python · 386 個 star

• RecML

  Python · 226 個 star

• xpk

  xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerat...

  Python · 193 個 star

• gpu-recipes

  Recipes for reproducing training and serving benchmarks for large machine learning models using GPUs on Google Cloud.

  Python · 140 個 star

• tpu-recipes

  Shell · 126 個 star

• accelerator-agents

  Python · 80 個 star

• ml-goodput-measurement

  Python · 41 個 star

• cloud-accelerator-diagnostics

  Python · 32 個 star

• accelerator-microbenchmarks

  Python · 24 個 star

• ray-tpu

  Python · 15 個 star

• resiliency

  This project provides resilient training logic to improve the training efficiency of large-scale distributed workloads on Google Cloud.

  Python · 14 個 star

• google-cloud-mldiagnostics

  Python · 14 個 star