Trending Projects

A deep learning package for many-body potential energy representation and molecular dynamics

A library focused on generating realistic talking head videos from images, useful for AI-powered avatar and animation projects.

MotionGPT is a unified motion-language generation model that can generate human motion using large language models.

A Python library for deep and online learning with spiking neural networks (SNNs) using PyTorch.

Flux 2 is a pure C inference engine for an image generation model, useful for vibe coders working with AI tools.

Text Generator is an Obsidian plugin that allows you to generate text content using various AI providers.

A Python library for creating invisible watermarks on images, useful for protecting digital content.

A secure low-code honeypot framework that leverages AI for system virtualization and preemptive cybersecurity.

A project to build a knowledge graph and AI agent for the renowned Chinese poet Li Bai, promoting his cultural legacy through AI-powered tools.

A playground for running AI red teaming exercises, including infrastructure setup.

A set of pre-trained machine learning models for the spaCy NLP library.

A collection of coding examples using the GPT-5 language model for various AI-powered coding tasks.

A comprehensive music transcription library that can detect beat, chord, drum, vocal, and instrument components.

LongWriter is a fine-tuned large language model (LLM) that can generate high-quality long-form text of 10,000+ words from long-form context.

An open-source driver assistance system that supports over 350 car makes and models.

An LLM compiler that enables efficient parallel function calling for large language models.

Repository containing human preference data for training a helpful and harmless AI assistant.

A research project from Facebook that explores multimodal AI models for computer vision and language tasks.

Open-source AI-powered computer use platform built with LLMs and a desktop sandbox

A dataset of worldwide building footprints derived from satellite imagery for geospatial and computer vision applications.

SCUDA is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.

A Python library for deep reinforcement learning in finance, with dynamic datasets and market environments.

A C++ library for tightly-coupled lidar-visual-inertial odometry and mapping, useful for robotics and autonomous systems development.

A Python library that projects the motion of pixels to a voxel representation, useful for vibe coders working with AI tools.

Cutting-edge AI-powered audio transcription tool with a user-friendly GUI and support for speaker identification.

PPL Quantization Tool (PPQ) is a powerful offline neural network quantization tool.

A Rust library and CLI tool for extracting text from images using optical character recognition (OCR).

A TypeScript-based financial agent for investment research and analysis.

Simple, hackable offline speech-to-text tool using the VOSK-API, useful for vibe coders building AI apps.

Versatile audio super resolution tool that can upscale audio to 48kHz using AI.

A package for the sparse identification of nonlinear dynamical systems from data

A platform for building and training multi-agent reinforcement learning systems.

A project demonstrating GPU-accelerated AI solutions for Lidar and camera data processing.

A platform for experimenting and researching the interaction of automated agents in simulated network environments.

A PyTorch extension library with optimized scatter operations for efficient data processing.

A deep learning-based interest network that discovers and connects developers with similar interests and coding styles.

A scalable, fast, and efficient approximate nearest neighbor search library written in Rust.

SEAL is a self-adapting language model that can continuously learn and improve its performance on various NLP tasks.

A free, private DeepSeek RAG chatbot with hybrid retrieval, neural reranking, and HyDe for boosting chatbot performance.

RLBench is a large-scale benchmark and learning environment for reinforcement learning agents.

A PyTorch library for training state-of-the-art multimodal multi-task models at scale.

An open-source project focused on building a large language model that understands human emotions and relationships, with tutorials on prompt engineering, RAG, agents, and LLM fine-tuning.

A comprehensive guide to implementing large language models with LangChain for AI/GenAI applications.

Powerful, fast, and efficient unstructured data extraction library written in Rust with language bindings.

JVector is an advanced embedded vector search engine written in Java, optimized for similarity search and machine learning applications.

Comprehensive resources for in-context learning and prompt engineering for large language models like ChatGPT and GPT-3.

Implements the 4 agentic patterns from scratch in a Jupyter Notebook for developers building with AI tools.

A flexible, high-performance e-graph library written in Rust for developers working with AI tools.

Underthesea is a powerful Vietnamese NLP toolkit for developers working with natural language processing tasks.

This is a Python library for audio tagging using a convolutional neural network model trained on the AudioSet dataset.