Trending Projects

SadTalker is a CVPR 2023 project that enables stylized audio-driven single image talking face animation.

A hierarchical reasoning model for large language models, focused on AI-powered developer tools and productivity.

Python library for invisible image watermarking that survives compression and transformations

Large Language Model Text Generation Inference library for developers working with AI tools and models.

Postgres-based database with GPU acceleration for machine learning and AI applications.

Personal website and blog of a developer focused on computer vision and machine learning.

MTEB is a benchmark for evaluating and comparing text embedding models across multiple tasks and languages.

A library for single- and multi-modal speaker verification, recognition, and diarization.

An effective paradigm for building tiny-scale vision-language-action models for robotics and embodied AI.

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs in Python.

An AI-powered tool to generate knowledge graphs from text data, with visualization capabilities.

A realistic web environment for building autonomous agents and AI-driven applications.

NVIDIA DLSS is a deep learning neural network that boosts frame rates and generates sharp images for games.

A starter kit for building AI agents on Cloudflare's serverless platform using TypeScript.

A Python library for generating human-centric videos using collaborative multi-modal conditioning.

One-click face swap tool for developers

Go client for OpenAI's ChatGPT, GPT-5, DALL-E, and Whisper APIs, enabling AI-powered applications in Go.

A Retrieval Augmented Generation (RAG) chatbot powered by Weaviate, a modern vector database for AI applications.

Flax is a flexible neural network library for JAX, designed for easy experimentation and research in machine learning.

A high-performance, AI-native database for LLM applications with hybrid search capabilities.

Latitude is an open-source platform for building, evaluating, and refining prompts for large language models.

FoundationStereo is a CVPR 2025 Best Paper Nomination project for zero-shot stereo matching using AI.

An open-source quantitative trading platform powered by reinforcement learning for finance and fintech developers.

A TypeScript library for a voice activity detector (VAD) with a simple API for the browser.

A comprehensive collection of tools, papers, and code for detecting deepfake media.

An open-source Python framework for implementing the Chanlun technical analysis methodology, supporting trading strategies and visualizations.

A real-time game translator with OCR that allows developers to build translation features into their games.

Envoy-based API gateway that manages unified access to generative AI services like GPT, DALL-E, and Stable Diffusion.

Distributed compiler based on Triton for parallel systems, focused on AI and high-performance computing.

A modular framework for robot learning from demonstration, focusing on AI-powered robotics and robotic simulation.

A tool for creating comics using AI, supporting script writing, storyboarding, and character style control.

A high-performance latent diffusion model for generating high-resolution images.

Kornia is a Python library for geometric computer vision and spatial AI tasks.

Pre-trained text-to-speech models for various languages, made simple to use.

Official implementation of DeepLabCut, a markerless pose estimation toolkit for animal behavior analysis using deep learning.

An open-source project that provides a state-of-the-art image restoration model using the Swin Transformer architecture.

A visual-inertial calibration toolbox for cameras and IMUs, enabling precise sensor fusion.

ML library for extracting metadata & text from PDFs using CRF & deep learning

A Python-based emotional companionship program powered by large language models (LLMs) for building AI-driven chatbots and virtual characters.

A digital avatar conversational system that combines large language models with visual models for novel human-AI interaction.

An open-source API management and distribution system that supports multiple AI models, including OpenAI.

A modular simulation framework and benchmark for robot learning using reinforcement learning and physics simulation.

An open-source text-to-speech tool supporting long-form text and multi-voice narration.

A comprehensive list of papers, code, and resources related to NeRF and 3D Gaussian Splatting for SLAM/Robotics applications.

An open-source library for automatically rigging diverse skeletons using a single neural network model.

A CVPR 2025 video diffusion model that enables fast autoregressive video generation from slow bidirectional models.

Bend is a massively parallel, high-level programming language written in Rust for building AI-powered applications.

FramePack is a Python library that makes video diffusion more practical, allowing developers to create AI-powered video editing tools.

A comprehensive list of PyTorch-related content on GitHub, including models, libraries, and tutorials.

Open-source script for improving object detection models, useful for vibe coders building AI-powered apps.