Trending Projects

Segment Anything Model for image/video segmentation with AI prompts

A PyTorch implementation of a CIFAR10 model achieving 95.47% accuracy.

JAX is a high-performance numerical computing library for Python, enabling transformations like differentiation and JIT compilation for GPUs/TPUs.

Flutter development tools for vibe coders

Isaac Lab API, powered by MuJoCo-Warp, for reinforcement learning and robotics simulation research.

A Python framework for high-performance auto-regressive diffusion model-based image and video generation.

VideoX is a collection of video cross-modal models for developers working with AI-powered video tools.

An AI-powered smart novel creation assistant to help you easily craft captivating stories.

JAX-based, hardware accelerated, batchable and differentiable optimizers for machine learning.

A PoC for a technique to recover plaintext from pixelized screenshots.

CoreNLP is a comprehensive NLP toolkit that provides powerful language processing capabilities for Java developers.

A WeChat bot powered by AI services like ChatGPT, Claude, and Kimi to automate various messaging tasks.

An implementation of the YOLOv9 object detection model, a state-of-the-art computer vision AI.

EmotiVoice is a multi-voice and prompt-controlled TTS engine built with PyTorch for developers working with AI voice tools.

Demos for the Claude AI agent SDK, enabling developers to build AI-powered applications and tools.

A fast BPE tokenizer for use with OpenAI's language models in Python.

ToonCrafter is a Python library for generative cartoon interpolation, presented at SIGGRAPH Asia 2024.

Notebooks for learning deep learning, focused on AI and machine learning

A fast, lightweight deep learning framework used in Alibaba's business-critical use cases, supporting LLM and 3D avatar apps.

A Python framework for creating AI agents that can learn to play any game you own

AgentScope Java: Agent-Oriented Programming framework for building LLM-powered applications

Open-source and modular AI-powered speech-to-speech translation tool built with Python.

One-click AI-powered short video creation and editing tool for product marketing and content

An open-source template that quickly sets up a local AI environment with essential tools for creating secure, self-hosted AI workflows.

Materials for learning PyTorch, a deep learning framework, from zero to mastery.

Fully automated AI video subtitle team with one-click subtitle cutting, translation, alignment, and dubbing.

A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthropic Claude models.

A Python library for serving large language models (LLMs) with high performance, including GPU acceleration and distributed inference.

A highly configurable and extensible Rime input scheme for Chinese writing with AI-powered word suggestions.

A TypeScript SDK for building AI-powered web applications and tools

A semantic router system for deploying and managing a mixture of AI models at the cloud, data center, and edge.

A Flutter-based LLM chat client with support for mobile and desktop platforms.

Open-source Grok-1 model for local inference with JAX

Triton is a development repository for a domain-specific programming language and compiler focused on machine learning and AI workloads.

A Python library for global optimization using Gaussian processes, useful for AI/ML developers.

An open-source library for quantizing diffusion models to 4-bit precision, absorbing outliers through low-rank components.

A foundational AI agent model for building agent-based, reasoning, and coding capabilities.

An open-source on-device speech recognition library for Apple Silicon devices, built with Swift and Transformers.

A high-performance Go engine for the game of Go, modeled after the AlphaGo Zero paper.

A fast, on-device, multilingual text-to-speech (TTS) library running natively via ONNX.

LLM training in C/CUDA with no dependencies

StyleGAN2 is an official TensorFlow implementation of a state-of-the-art generative adversarial network.

An interactive visualization tool that explains how transformer language models work, targeting AI-focused developers.

PyTorch implementation of Graph Convolutional Networks, a powerful machine learning technique for graph data.

A reverse-engineered Python API for interacting with the Google Gemini web app, a generative AI tool.

Open-source platform for building enterprise-grade agents with RAG, workflows, and MCP tools

This project aims to speed up large language model (LLM) inference and enhance their understanding of key information through prompt and KV-Cache compression.

A community-driven platform that transforms study materials into interactive resources like quizzes, flashcards, notes, and podcasts.

An open-source deep research agent from Alibaba that can assist with information seeking and AI tasks.

A flexible, high-performance deep learning framework for Python that runs on GPUs.