Stars
Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation
Annotated version of the Mamba paper
Ongoing research training transformer models at scale
depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.
What would you do with 1000 H100s...
Easily train a good VC model with voice data <= 10 mins!
Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.
deepspeedai / Megatron-DeepSpeed
Forked from NVIDIA/Megatron-LMOngoing research training transformer language models at scale, including: BERT & GPT-2
CLU lets you write beautiful training loops in JAX.
Minimal library to train LLMs on TPU in JAX with pjit().
A playbook for systematically maximizing the performance of deep learning models.
Wayback Machine API interface & a command-line tool
Various Jupyter notebooks about Common Crawl data
Fast Inference Solutions for BLOOM
Machine Learning Engineering Open Book
Platform for building statistical models of cities and regions
Hostfile blocklist for ads and tracking, updated regularly