🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

FastAPI framework, high performance, easy to learn, fast to code, ready for production

Robust Speech Recognition via Large-Scale Weak Supervision

Python tool for converting files and office documents to Markdown.

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), ga…

Deep Learning for humans

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Official Code for DragGAN (SIGGRAPH 2023)

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Convert PDF to markdown + JSON quickly with high accuracy

You like pytorch? You like micrograd? You love tinygrad! ❤️

Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 500 universities from 70 countries including Stanford, MIT, Harvard, and Cambridge.

🍰 Desktop utility to download images/videos/music/text from various websites, and more.

Documentation that simply works

Python logging made (stupidly) simple

A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.

Mamba SSM architecture

pix2tex: Using a ViT to convert images of equations into LaTeX code.

PyGWalker: Turn your dataframe into an interactive UI for visual analysis

A computer algebra system written in pure Python

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 RN…

Wan: Open and Advanced Large-Scale Video Generative Models

PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

Minimal reproduction of DeepSeek R1-Zero

Hierarchical Reasoning Model Official Release

Source code for Twitter's Recommendation Algorithm

Code for the manim-generated scenes used in 3blue1brown videos

Streamlined interface for generating images with AI in Krita. Inpaint and outpaint with optional text prompt, no tweaking required.

Simultaneous speech-to-text model