Trending AI projects

DeerFlow 二次开发：理论、架构与源码剖析 is a community-driven, MIT-licensed book that explains the internal source code of DeerFlow 2.0. It helps developers move from using the framework to customizing it, covering architecture, Skills, Sub-Agents, Sandbox, Memory, MCP integration, and enterprise patterns, with a Docker quick-start and practical examples.

A curated collection of prompts and examples designed to help users master OpenAI's GPT Image 2 model for high-fidelity image generation.

A Spec-Driven Development tool consisting of a desktop app, CLI, and AI agent skills to visually track and manage specifications, change proposals, and tasks.

AI-Infra provides a visual landscape and organized learning path for AI infrastructure engineers, focusing on Kubernetes and cloud-native environments. It guides users through essential components and advanced topics, with links to detailed documentation and community resources.

imsg is a Swift-based command-line tool for macOS that enables reading, watching, and sending iMessage and SMS messages through the terminal, providing JSON output for easy automation and integration.

An open GitHub repository collecting free course materials—notes, past exams, and study guides—for students in the AI School at Anhui University, meant to combat the selling of learning resources and ensure equal access for all learners.

UltraBr3aks is a repository that shares prompt-based jailbreak techniques for major AI models like GPT, Claude, and Gemini. It documents various attack methods, such as Attention-Breaking and token injection, to help researchers study and understand LLM safety vulnerabilities.

A practical, framework-agnostic book that teaches AI agent architecture patterns from basic ReAct loops to enterprise multi-agent orchestration, with a reference implementation (Shannon) spanning Go, Rust, and Python.

Codex is a programming agent that helps users manage and execute technical projects by acting as an autonomous assistant that can plan, code, and verify tasks. It bridges the gap for non-experts by providing a structured, agentic environment for building software and automating workflows.

A comprehensive Chinese tutorial project for Claude Code, offering over 212 articles and 29 structured chapters that take learners from basic installation to enterprise-level workflows, including MCP, skills, plugins, and multi-agent collaboration.

A curated index of autonomous AI research agents and self-improvement loops that automate tasks like code optimization, scientific discovery, and metric-driven experimentation.

A collection of up-to-date, tested code examples and tutorials for building AI agents using frameworks like AutoGen, CrewAI, and PydanticAI.

A native macOS command center and terminal designed for running and coordinating multiple AI coding agents in parallel, featuring state tracking and multi-agent workflows.

EasyJailbreak is an open‑source Python framework that modularises LLM jailbreak attacks. It provides ready‑made recipes (ReNeLLM, GPTFuzz, AutoDAN, etc.) and lets users plug in custom selectors, mutators, constraints, and evaluators to generate, run, and score adversarial prompts against target language models. Install via `pip install easyjailbreak` or from source, then use the high‑level API (e.g., `PAIR` recipe) or build your own attack pipeline. The project includes documentation, a paper, datasets, and downloadable experimental results.

A pure C implementation of an AI assistant for the ESP32-S3 chip, providing a low-power, always-on agent with local memory and tool-calling capabilities via Telegram.

Ersilia is an open‑source CLI and Python library that lets you fetch, serve, and run pre‑trained AI/ML models for drug‑discovery (e.g., antibiotic activity, ADMET). Models are packaged in isolated Docker/Conda environments and accessed via simple commands (`ersilia fetch`, `serve`, `run`). The tool targets infectious and neglected disease research, offering a low‑code way to apply curated models without handling complex dependencies.

Integuru v0 is an AI agent that reverse-engineers browser network requests to generate runnable Python code for platforms without official APIs.

Isaac Capture is NVIDIA’s unified stack for recording robot tele‑operation and egocentric human‑centric data. It bundles the Isaac Teleop interface (XR, glove, gamepad control of robots) with an Ego Pipeline that reconstructs hand and camera poses from head‑mounted video. Shared device plugins, retargeting utilities, and data handling work in both simulation (Isaac Lab) and real‑world robots, making it a ready‑to‑use foundation for learning‑from‑demonstration, tele‑operation research, and vision dataset creation.

A scientific machine learning library for solving differential equations and learning operators using physics-informed neural networks and other advanced AI architectures.

An MCP server that enables AI agents to securely manage remote servers via SSH, featuring per-server security modes to control command execution.

Mini Pupper 4 is an open-source quadruped robotics platform that enables developers to train neural network policies in high-fidelity simulations and deploy them directly to physical hardware using ROS 2 and micro-ROS.

FlowDown is a native, privacy-focused AI chat client for Apple platforms that allows users to connect to local or remote AI models while keeping all data on their device or private iCloud.

A community-curated list of AI-assisted coding tools and articles, organized by category — web builders, editors, CLI tools, plugins, and more — centered on the “vibe coding” approach of developing software through prompts and iteration.

A set of plugins for Claude Code and Codex that uses evidence-based learning science to provide interactive exercises, preventing skill decay during AI-assisted coding.

A framework that turns coding models into browser and desktop agents by giving them a terminal to write and execute automation scripts.

UniRL is an Apache‑2.0 Python framework that applies a single reinforcement‑learning loop to many multimodal generative models (diffusion, video, LLM, VLM, and unified AR+diffusion). It offers composable trainers, multiple rollout back‑ends (vLLM‑Omni, SGLang), standard and novel RL algorithms (GRPO, Flow‑DPPO, DRPO, CPPO), and ready‑to‑run recipes for dozens of state‑of‑the‑art models. Users launch training via Hydra configs, and the system handles distributed rollout, scoring, advantage computation, and weight syncing.

A comprehensive multi-agent framework and educational resource that teaches how to build, orchestrate, and evaluate AI agent systems from first principles.

An open-source observability framework built on OpenTelemetry that provides complete tracing and monitoring for LLM applications, providers, and vector databases.

A cross-platform visual database schema designer that integrates with IDEs and allows AI coding agents to collaborate on schema design via the Model Context Protocol.

An LSPosed/Xposed module for the official DeepSeek Android app that provides a compatibility layer and additional settings to enhance the mobile user experience.

A comprehensive firmware for ESP32-CAM modules that enables motion-triggered or continuous video recording to SD cards, live streaming, and integration with home automation systems.

Ibis is a portable Python dataframe library that provides a unified API for interacting with over 20 different data backends, allowing users to write code once and deploy it across various SQL and dataframe engines.

autoskills is a zero-config CLI tool that scans your project's files (package.json, Gradle, configs) to detect your tech stack, then downloads and verifies curated AI agent skills from a security-reviewed registry — with hashed manifests, AI review for supply-chain attacks, a lockfile, and options like dry-run and per-IDE installs.

A hands-on, 15-checkpoint educational course that guides developers through building a functional coding agent using real Git history and verifiable code.

An experimental tool that transforms unstructured documents and web content into structured Knowledge Graphs stored in Neo4j using LLMs, enabling conversational querying and graph visualization.

An AI-powered smart resume builder that offers real-time AI optimization, JD matching, and mock interview simulations to help users land jobs.

A desktop application that provides a zero-install, GUI-based interface for translating academic PDFs while preserving complex formatting, formulas, and charts, with deep integration for Zotero.

A desktop AIOps agent for Windows and macOS that automates IT operations tasks like alert analysis and system inspection through natural language and toolchain integration.

A unified Python SDK that provides a single interface to communicate with multiple LLM providers, allowing developers to switch models without changing their code.

An extension of the Hugging Face ecosystem that provides optimization tools to train and run AI models with maximum efficiency on targeted hardware accelerators.

A tool and SDK that lets you use your ChatGPT account as an OpenAI-compatible API, enabling developers to build apps where users bring their own AI accounts.

Python bindings for llama.cpp that enable local LLM inference with an OpenAI-compatible API and support for multiple hardware acceleration backends.

A collection of tutorials and reference implementations for AgentKit, an enterprise AI Agent platform, covering everything from basic dialogue to complex multi-agent systems and RAG.

An RLM-based agent optimizer that analyzes production execution traces to identify systemic failure modes and provide recommendations for improving agent harnesses.

A tiny, agent-native CLI for generating text, images, video, and audio using the Vercel AI SDK and AI Gateway for unified model access.

bae is a PyTorch library providing efficient, exact second-order sparse optimization algorithms for robotics, including Bundle Adjustment and Pose Graph Optimization. It leverages CUDA acceleration and integrates with PyTorch's autograd for high-performance non-linear least squares problems.

A Python toolkit for RAG research that provides a modular framework, 23 SOTA algorithms, and 36 benchmark datasets to simplify the reproduction and development of retrieval-augmented generation systems.

SpecStory is an open‑source CLI (with IDE extensions) that automatically records every conversation you have with AI coding assistants, stores them locally, and optionally syncs them to a cloud service for searchable, shareable knowledge. It supports many agents (Cursor, Copilot, Claude, Codex, etc.) and includes a “Lore” feature that turns saved sessions into reusable agent skills.