A high-throughput and memory-efficient inference and serving engine for LLMs

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

Pyodide is a Python distribution for the browser and Node.js based on WebAssembly

Hackable and optimized Transformers building blocks, supporting a composable construction.

The official PyTorch implementation of Google's Gemma models

A library for mechanistic interpretability of GPT-style language models

OpenAI-Compatible RESTful APIs for Amazon Bedrock