AI-Hypercomputer / torchprimeLinks
torchprime is a reference model implementation for PyTorch on TPU.
☆39Updated last week
Alternatives and similar repositories for torchprime
Users that are interested in torchprime are comparing it to the libraries listed below
Sorting:
- ☆15Updated 5 months ago
- ☆121Updated last year
- ☆14Updated 5 months ago
- ☆190Updated this week
- A library for unit scaling in PyTorch☆132Updated 3 months ago
- Load compute kernels from the Hub☆308Updated this week
- Scalable and Performant Data Loading☆311Updated last week
- JAX bindings for Flash Attention v2☆97Updated last week
- Accelerated First Order Parallel Associative Scan☆189Updated last year
- ☆335Updated last month
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆270Updated 3 months ago
- ☆542Updated last year
- A set of Python scripts that makes your experience on TPU better☆54Updated last month
- JAX implementation of the Llama 2 model☆216Updated last year
- Minimal yet performant LLM examples in pure JAX☆187Updated last month
- PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"☆76Updated last month
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"☆241Updated 4 months ago
- ☆91Updated last year
- ☆149Updated 2 years ago
- Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax☆671Updated this week
- seqax = sequence modeling + JAX☆168Updated 3 months ago
- This repository contains the experimental PyTorch native float8 training UX☆223Updated last year
- ☆145Updated 2 weeks ago
- Google TPU optimizations for transformers models☆121Updated 9 months ago
- Two implementations of ZeRO-1 optimizer sharding in JAX☆14Updated 2 years ago
- Accelerate, Optimize performance with streamlined training and serving options with JAX.☆317Updated this week
- Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimenta…☆539Updated last month
- ☆283Updated last year
- Inference code for LLaMA models in JAX☆119Updated last year
- ☆23Updated last week