Starred repositories
On-device AI across mobile, embedded and edge for PyTorch
High-performance In-browser LLM Inference Engine
Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.
A Python implementation of Tensor Logic - a unified programming language for AI that combines neural and symbolic reasoning through tensor equations.
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RN…
Unofficial implementation of Titans, SOTA memory for transformers, in Pytorch
AgentEvolver: Towards Efficient Self-Evolving Agent System
SQL Native Memory Layer for LLMs, AI Agents & Multi-Agent Systems
Memory infrastructure for LLMs and AI agents
Tiny and powerful JavaScript full-text search engine for browser and Node
Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning" by Zhiheng Xi et al.
Code and implementations for the ACL 2025 paper "AgentGym: Evolving Large Language Model-based Agents across Diverse Environments" by Zhiheng Xi et al.
Efficent platform for inference and serving local LLMs including an OpenAI compatible API server.
A web-based collaborative LaTeX editor
🗂️A file list/WebDAV program that supports multiple storages, powered by Gin and Solidjs. / 一个支持多存储的文件列表/WebDAV程序，使用 Gin 和 Solidjs。
An open-source remote desktop application designed for self-hosting, as an alternative to TeamViewer.
Batfish is a network configuration analysis tool that can find bugs and guarantee the correctness of (planned or current) network configurations. It enables network engineers to rapidly and safely …
AgentScope: Agent-Oriented Programming for Building LLM Applications
Master programming by recreating your favorite technologies from scratch.
Multilingual Document Layout Parsing in a Single Vision-Language Model
Cambrian-1 is a family of multimodal LLMs with a vision-centric design.
An API standard for multi-agent reinforcement learning environments, with popular reference environments and related utilities
High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)
LLM agents built for control. Designed for real-world use. Deployed in minutes.
The C++ Core Guidelines are a set of tried-and-true guidelines, rules, and best practices about coding in C++
Towards a Unified View of Large Language Model Post-Training