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

Mounts Media Transfer Protocol (MTP) devices as (removable) drives on Windows for access via command line.

VPTQ, A Flexible and Extreme low-bit quantization algorithm

Investment Research for Everyone, Everywhere.

Implementing a ChatGPT-like LLM in PyTorch from scratch, step by step

Tensor parallelism is all you need. Run LLMs on an AI cluster at home using any device. Distribute the workload, divide RAM usage, and increase inference speed.

AI Logging for Interpretability and Explainability🔬

🙌 OpenHands: Code Less, Make More

GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

open-source code for paper: Retrieval Head Mechanistically Explains Long-Context Factuality

Cross-Platform, GPU Accelerated Whisper 🏎️

Llama-3 agents that can browse the web by following instructions and talking to you

Code for the ProteinMPNN paper

[ICML'24 Spotlight] LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning

💢 Pressure testing the context window of open LLMs

The official Meta Llama 3 GitHub site

Streisand sets up a new server running your choice of WireGuard, OpenConnect, OpenSSH, OpenVPN, Shadowsocks, sslh, Stunnel, or a Tor bridge. It also generates custom instructions for all of these s…

[ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning

Popcorn Time is a multi-platform, free software BitTorrent client that includes an integrated media player ( Windows / Mac / Linux ) A Butter-Project Fork

Mamba SSM architecture

Open-source vector similarity search for Postgres

DSPy: The framework for programming—not prompting—foundation models

A simple "Be My Eyes" web app with a llama.cpp/llava backend

Real-time face swap for PC streaming or video calls

real time face swap and one-click video deepfake with only a single image