AI-Hypercomputer / kitharaLinks
☆14Updated last month
Alternatives and similar repositories for kithara
Users that are interested in kithara are comparing it to the libraries listed below
Sorting:
- torchprime is a reference model implementation for PyTorch on TPU.☆28Updated last week
- ☆141Updated 3 weeks ago
- ☆34Updated last week
- PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"☆62Updated 2 months ago
- ☆21Updated last week
- ☆16Updated 3 months ago
- Recipes for reproducing training and serving benchmarks for large machine learning models using GPUs on Google Cloud.☆73Updated last week
- A simplified and automated orchestration workflow to perform ML end-to-end (E2E) model tests and benchmarking on Cloud VMs across differe…☆48Updated this week
- JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs wel…☆349Updated 2 weeks ago
- Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)☆64Updated last week
- xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerat…☆125Updated this week
- Google TPU optimizations for transformers models☆113Updated 5 months ago
- ☆186Updated 3 weeks ago
- PyTorch Single Controller☆231Updated this week
- ☆504Updated 11 months ago
- Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimenta…☆510Updated last week
- ☆17Updated last week
- Inference code for LLaMA models in JAX☆118Updated last year
- Two implementations of ZeRO-1 optimizer sharding in JAX☆14Updated 2 years ago
- ☆126Updated last month
- A JAX-native LLM Post-Training Library☆58Updated this week
- ☆87Updated last week
- PyTorch per step fault tolerance (actively under development)☆329Updated this week
- ☆318Updated this week
- Orbax provides common checkpointing and persistence utilities for JAX users☆393Updated this week
- ☆67Updated 2 years ago
- My annotated papers and meeting recordings for the EleutherAI ML Performance research paper reading group☆13Updated last month
- Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax☆596Updated this week
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆179Updated 2 weeks ago
- Experimenting with how best to do multi-host dataloading☆10Updated 2 years ago