AI-Hypercomputer / xpk
xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerators such as TPUs and GPUs on GKE.
☆110Updated this week
Alternatives and similar repositories for xpk:
Users that are interested in xpk are comparing it to the libraries listed below
- JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs wel…☆313Updated this week
- ☆137Updated this week
- Recipes for reproducing training and serving benchmarks for large machine learning models using GPUs on Google Cloud.☆53Updated last week
- Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimenta…☆488Updated this week
- A simplified and automated orchestration workflow to perform ML end-to-end (E2E) model tests and benchmarking on Cloud VMs across differe…☆43Updated this week
- Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)☆64Updated 4 months ago
- ☆38Updated 2 weeks ago
- ☆20Updated this week
- AI on GKE is a collection of examples, best-practices, and prebuilt solutions to help build, deploy, and scale AI Platforms on Google Kub…☆302Updated this week
- PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"☆59Updated 2 weeks ago
- ☆186Updated this week
- PyTorch per step fault tolerance (actively under development)☆274Updated this week
- ☆200Updated this week
- ☆19Updated this week
- Google TPU optimizations for transformers models☆107Updated 2 months ago
- ☆419Updated 9 months ago
- ☆43Updated 2 months ago
- Orbax provides common checkpointing and persistence utilities for JAX users☆364Updated this week
- ☆16Updated last month
- JAX-Toolbox☆298Updated this week
- Module, Model, and Tensor Serialization/Deserialization☆221Updated last month
- ☆190Updated last week
- ☆15Updated last week
- ☆295Updated this week
- JAX implementation of the Llama 2 model☆217Updated last year
- seqax = sequence modeling + JAX☆153Updated last week
- Library for reading and processing ML training data.☆425Updated this week
- TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and sup…☆358Updated last week
- ☆67Updated 2 years ago
- Kubernetes Operator, ansible playbooks, and production scripts for large-scale AIStore deployments on Kubernetes.☆92Updated this week