Madison May

nono. Sandbox any AI agent in seconds - zero setup, zero latency.

harmony. Renderer for the harmony response format to be used with gpt-oss

langextract. A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

AdalFlow. AdalFlow: The library to build & auto-optimize LLM applications.

zerox. OCR & Document Extraction using vision models

phoenix. AI Observability & Evaluation

attention-gym. Helpful tools and examples for working with flex-attention

ThunderKittens. Tile primitives for speedy kernels

HeavyBall. Efficient optimizers

bayesflow. A Python library for efficient Bayesian modeling with deep learning

unstract-sdk. A framework for writing Unstract Tools/Apps

docai. Structured information extraction from documents

byaldi. Use late-interaction multi-modal models such as ColPali in just a few lines of code.

llm_aided_ocr. Enhances Tesseract OCR output using LLMs (local or API) for error correction, smart chunking, and markdown formatting of scanned PDFs

six-degrees-of-torvalds. TypeScript

Devon. Devon: An open-source pair programmer

full-stack-fastapi-template. Full stack, modern web application template. Using FastAPI, React, SQLModel, PostgreSQL, Docker, GitHub Actions, automatic HTTPS and more.

unitable. UniTable: Towards a Unified Table Foundation Model

amalgam-lang-py. Amalgam Python Wrapper

fuxion. Sythetic data generation and normalization functions powered by LLMs

lorax. Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

surya. OCR, layout analysis, reading order, table recognition in 90+ languages

pdfminer.six. Community maintained fork of pdfminer - we fathom PDF

daytona. Daytona is a Secure and Elastic Infrastructure for Running AI-Generated Code

nanodl. JAX library for training sub-4B foundation models for edge

RAGatouille. Easily use and train state of the art late-interaction retrieval methods (ColBERT) in any RAG pipeline. Designed for modularity and ease-of-use, backed by research.

outlines. Structured Outputs

litgpt. 20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

Doc2Dict. Code accompanying Doc2Dict paper

autolabel. Label, clean and enrich text datasets with LLMs.

simple-hierarchical-transformer. Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT

MEGABYTE-pytorch. Implementation of MEGABYTE, Predicting Million-byte Sequences with Multiscale Transformers, in Pytorch

LlamaAcademy. A school for camelids

AutoGPT. AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

babyagi. Python

the-algorithm. Source code for the X Recommendation Algorithm

OpenChatKit. Python

FLAN. Python

llm-security. New ways of breaking app-integrated LLMs

dolly. Databricks’ Dolly, a large language model trained on the Databricks Machine Learning Platform

chat-langchain. TypeScript

pal. PaL: Program-Aided Language Models (ICML 2023)

recursive_llm. Implement recursion using English as the programming language and an LLM as the runtime.

ThoughtSource. A central, open resource for data and tools related to chain-of-thought reasoning in large language models. Developed @ Samwald research group: https://samwald.info/

xsdata. Naive XML & JSON Bindings for python

llama_index. LlamaIndex is the leading document agent and OCR platform

langchain. The agent engineering platform.

cramming. Cramming the training of a (BERT-type) language model into limited compute.

point-e. Point cloud diffusion for 3D model synthesis

awesome-document-understanding. A curated list of resources for Document Understanding (DU) topic

bistring. Bidirectionally transformed strings

PaddleOCR. Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.

ocrpy. OCR, Archive, Index and Search: Implementation agnostic OCR framework.

azure-docs. Open source documentation of Microsoft Azure

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

flash-attention. Fast and memory-efficient exact attention

perceiver-ar-pytorch. Implementation of Perceiver AR, Deepmind's new long-context attention network based on Perceiver architecture, in Pytorch

nlp. Python

ema-pytorch. A simple way to keep track of an Exponential Moving Average (EMA) version of your Pytorch model

silero-models. Silero Models: pre-trained text-to-speech models made embarrassingly simple

DeBERTa. The implementation of DeBERTa

flamingo-pytorch. Implementation of 🦩 Flamingo, state-of-the-art few-shot visual question answering attention net out of Deepmind, in Pytorch

imagen-pytorch. Implementation of Imagen, Google's Text-to-Image Neural Network, in Pytorch

DIMetrics. Jupyter Notebook

qdrant. Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/

doubtlab. Doubt your data, find bad labels.

indico-client-python. Indico IPA client library

dalle-flow. 🌊 A Human-in-the-Loop workflow for creating HD images from text

equinox. Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/

jaxton. 100 exercises to learn JAX

keops. KErnel OPerationS, on CPUs and GPUs, with autodiff and without memory overflows

gradio. Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

rich-cli. Rich-cli is a command line toolbox for fancy output in the terminal

performer-pytorch. An implementation of Performer, a linear attention-based transformer, in Pytorch

HiddenEngrams. Hidden Engrams: Long Term Memory for Transformer Model Inference

isab-pytorch. An implementation of (Induced) Set Attention Block, from the Set Transformers paper

hippo-code. Python

label-studio. Label Studio is a multi-type data labeling and annotation tool with standardized output format

triton-transformer. Implementation of a Transformer, but completely in Triton

autofaiss. Automatically create Faiss knn indices with the most optimal similarity search parameters.

Transformers-Tutorials. This repository contains demos I made with the Transformers library by HuggingFace.

vae-textures. Texture mapping with variational auto-encoders

milvus. Milvus is a high-performance, cloud-native vector database built for scalable vector ANN search

doctr. docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.

fast-transformer-pytorch. Implementation of Fast Transformer in Pytorch

MathBox.js. MathBox is a (work in progress) library for making presentation-quality math diagrams in WebGL.

lightly. A python library for self-supervised learning on images.

annotated_deep_learning_paper_implementations. 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

ALBEF. Code for ALBEF: a new vision-language pre-training method

ide. IDE for Lark-Parser

FastAI.jl. Repository of best practices for deep learning in Julia, inspired by fastai

Metroplex. Python

OBST. Your fruity companion for transformers

dex-lang. Research language for array processing in the Haskell/ML family

Multimodal-Toolkit. Multimodal model for text and tabular data with HuggingFace transformers as building block for text data

mindshub. Make AI do actual work. Swap the model anytime — keep everything you've built.

tagruler. Data programming by demonstration for information extraction and span annotation

pyprobml. Python code for "Probabilistic Machine learning" book by Kevin Murphy

lm-evaluation-harness. A framework for few-shot evaluation of language models.

equivariance. A framework for implementing equivariant DL

project-menu. See the issue board for the current status of active and prospective projects!

mesh-transformer-jax. Model parallel transformers in JAX and Haiku

fairseq-apollo. FairSeq repo with Apollo optimizer

textual. The lean application framework for Python. Build sophisticated user interfaces with a simple Python API. Run your apps in the terminal and a web browser.

redis-inference-optimization. A Redis module for serving tensors and executing deep learning graphs

dl-visuals. Over 200 figures and diagrams of the most popular deep learning architectures and layers FREE TO USE in your blog posts, slides, presentations, or papers.

loky. Robust and reusable Executor for joblib

g-mlp-pytorch. Implementation of gMLP, an all-MLP replacement for Transformers, in Pytorch

guided-diffusion. Python

x-transformers. A concise but complete full-attention transformer with a set of promising experimental features from various papers

gpt-neox. An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

dino. PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

DeOldify. A Deep Learning based project for colorizing and restoring old images (and video!)

torchtyping. Type annotations and dynamic checking for a tensor's shape, dtype, names, etc.

skweak. skweak: A software toolkit for weak supervision applied to NLP tasks

accelerate. 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

layout-parser. A Unified Toolkit for Deep Learning Based Document Image Analysis

blackjax. BlackJAX is a Bayesian Inference library designed for ease of use, speed and modularity.

big-sleep. A simple command line tool for text to image generation, using OpenAI's CLIP and a BigGAN. Technique was originally created by https://twitter.com/advadnoun

deep-daze. Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network). Technique was originally created by https://twitter.com/advadnoun

perceiver-pytorch. Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

PlotNeuralNet. Latex code for making neural networks diagrams

autotrain-advanced. 🤗 AutoTrain Advanced

DALL-E. PyTorch package for the discrete VAE used for DALL·E.

allie. 🤖 An automated machine learning framework for audio, text, image, video, or .CSV files (50+ featurizers and 15+ model trainers). Python 3.6 required.

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

DALLE-pytorch. Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

vit-pytorch. Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

tensor_annotations. Annotating tensor shapes using Python types

jax-md. Differentiable, Hardware Accelerated, Molecular Dynamics

ocr-connectionist-text-network. A end to end text detection pipeline using CTPN

TextRecognitionDataGenerator. A synthetic data generator for text recognition

CLEval. CLEval: Character-Level Evaluation for Text Detection and Recognition Tasks

TablaM. The practical relational programing language for data-oriented applications

neural_ca. Neural Cellular Automata

fastformers. FastFormers - highly efficient transformer models for NLU

jraph. A Graph Neural Network Library in Jax

checklist. Beyond Accuracy: Behavioral Testing of NLP models with CheckList

just-pandas-things. An ongoing list of pandas quirks

literature-human-ai-collaboration. Literature for Human-AI collaboration, Hybrid Intelligence, Human-AI interaction

lambda-networks. Implementation of LambdaNetworks, a new approach to image recognition that reaches SOTA with less compute

pydantic. Data validation using Python type hints

DeepSpeed. DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

human-learn. Natural Intelligence is still a pretty good idea.

concise-encoding. The secure data format for a modern world

labml. 🔎 Monitor deep learning model training and hardware usage from your mobile phone 📱

TextAttack. TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

pytorch_block_sparse. Fast Block Sparse Matrices for Pytorch

KILT. Library for Knowledge Intensive Language Tasks

applied-ml. 📚 Papers & tech blogs by companies sharing their work on data science & machine learning in production.

strux. Python

objax. Python

einops. Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

nlphighlights. Website for the NLP Highlights podcast, including transcriptions of each episode

slate. Beautiful static documentation for your API

optax. Optax is a gradient processing and optimization library for JAX.

deepmind-research. This repository contains implementations and illustrative code to accompany DeepMind publications

latexify_py. A library to generate LaTeX expression from Python code.

gpt-scrolls. A collaborative collection of open-source safe GPT-3 prompts that work well

elegy. A High Level API for Deep Learning in JAX

haystack. Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems.

adapters. A Unified Library for Parameter-Efficient and Modular Transfer Learning

cdx_toolkit. A toolkit for CDX indices such as Common Crawl and the Internet Archive's Wayback Machine

serve. ☁️ Build multimodal AI applications with cloud-native stack

local-attention. An implementation of local windowed attention for language modeling

fast-soft-sort. Fast Differentiable Sorting and Ranking

fast-transformers. Pytorch library for fast transformer implementations

gmat. The accompanying code for the paper "GMAT: Global Memory Augmentation for Transformers" (Ankit Gupta and Jonathan Berant).

OTK. A Pytorch implementation of the optimal transport kernel embedding

BlingFire. A lightning fast Finite State machine and REgular expression manipulation library.

compressive-transformer-pytorch. Pytorch implementation of Compressive Transformers, from Deepmind

siren. Official implementation of "Implicit Neural Representations with Periodic Activation Functions"

pyright. Static Type Checker for Python

caliban. Research workflows made easy, locally and in the Cloud.

practical-python. Practical Python Programming (course by @dabeaz)

byol-pytorch. Usable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch

triton. Development repository for the Triton language and compiler

torch-blocksparse. Block-sparse primitives for PyTorch

sha-rnn. Single Headed Attention RNN - "Stop thinking with your head"

mmf. A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

virtex. [CVPR 2021] VirTex: Learning Visual Representations from Textual Annotations

LSR. Pytorch Implementation of our ACL 2020 Paper "Reasoning with Latent Structure Refinement for Document-Level Relation Extraction"

product-key-memory. Standalone Product Key Memory module in Pytorch - for augmenting Transformer models

axial-positional-embedding. Axial Positional Embedding for Pytorch

Funnel-Transformer. Python

tvm. Open Machine Learning Compiler Framework

RSE. Residual Shuffle-Exchange

penrose. Create beautiful diagrams just by typing notation in plain text.

linear-attention-transformer. Transformer based on a variant of attention that is linear complexity in respect to sequence length

opt_einsum. ⚡️Optimizing einsum functions in NumPy, Tensorflow, Dask, and more with contraction order optimization.

gpt-3. GPT-3: Language Models are Few-Shot Learners

routing-transformer. Fully featured implementation of Routing Transformer

mathemagic. Toybox of explanations of mathematics. Initial focus on (discrete) differential geometry

parallax. Python

littleballoffur. Little Ball of Fur - A graph sampling extension library for NetworKit and NetworkX (CIKM 2020)

nn4nlp-concepts. A repository of concepts related to neural networks for NLP

cmu-multinlp. Generalizing Natural Language Analysis through Span-relation Representations

pet. This repository contains the code for "Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference"

contrastive-learner. A simple to use pytorch wrapper for contrastive self-supervised learning on any neural network

PubLayNet. Jupyter Notebook