Stars
Tutel MoE: Optimized Mixture-of-Experts Library, Support DeepSeek FP8/FP4
A toolkit for React, Preact, Inferno & vanilla JS apps, React libraries and other npm modules for the web, with no configuration (until you need it)
A Library for Differentiable Logic Gate Networks
Implementation of Muse: Text-to-Image Generation via Masked Generative Transformers, in Pytorch
Blazing fast macOS PHP development environment
Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparen…
Code for Switchable Normalization from "Differentiable Learning-to-Normalize via Switchable Normalization", https://arxiv.org/abs/1806.10779
🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading
Code for the ICML 2021 (long talk) paper: "ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision"
A lightweight library for PyTorch training tools and utilities
NumPy aware dynamic Python compiler using LLVM
A lightweight LLVM python binding for writing JIT compilers
🦜🔗 Build context-aware reasoning applications
Directly Connecting Python to LLMs via Strongly-Typed Functions, Dataclasses, Interfaces & Generic Types
Materials for the Learn PyTorch for Deep Learning: Zero to Mastery course.
🧑🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), ga…
Robust Speech Recognition via Large-Scale Weak Supervision
A temporary repository hosting a pomegranate re-write using PyTorch as the backend.
Fast, flexible and easy to use probabilistic modelling in Python.
A Python implementation of active inference for Markov Decision Processes
The full stack toolkit to build onchain app UX
Approximate nearest neighbor search with product quantization on GPU in pytorch and cuda
Transformer related optimization, including BERT, GPT
tiktoken is a fast BPE tokeniser for use with OpenAI's models.