tiktoken is a fast BPE tokeniser for use with OpenAI's models
A Python package to assess and improve fairness of ML models
Probabilistic reasoning and statistical analysis in TensorFlow
Library to facilitate federated learning research
Easy-to-use and high-performance NLP and LLM framework
ExtractThinker is a Document Intelligence library for LLMs
The no-nonsense RAG chunking library
The library to build & auto-optimize LLM applications
Explainability and Interpretability to Develop Reliable ML models
Python examples of popular machine learning algorithms
A Unified Library for Parameter-Efficient Learning
Build resilient language agents as graphs
High-Performance Face Recognition Library on PaddlePaddle & PyTorch
Advanced AI Explainability for computer vision
Accessible large language models via k-bit quantization for PyTorch
A Production-ready Reinforcement Learning AI Agent Library
A PyTorch library for implementing flow matching algorithms
PyTorch3D is FAIR's library of reusable components for deep learning
End-to-End Library for Continual Learning based on PyTorch
Converts text to speech in realtime
A unified framework for machine learning with time series
A very simple framework for state-of-the-art NLP
An open and fair framework for everyone to build AI agents
Sample code and notebooks for Generative AI on Google Cloud
Prevent PyTorch's `CUDA error: out of memory` in just 1 line of code