Trending AI projects

Surya is a 650M-parameter OCR model that turns scanned documents into text plus structure — layout blocks, reading order, table rows and columns, and inline math as LaTeX. It leads benchmarks for its size (83.3% on olmOCR-bench), supports 91 languages, and runs about 5 pages per second on a modern GPU.

Project SuperDex is a unified simulation platform for dexterous manipulation research, providing a physics engine, robotics SDK, and reinforcement learning interface in one integrated environment.

An intent compiler for AI agents that automatically probes your codebase and documentation to resolve ambiguities, asking only the necessary questions and saving decisions as persistent, verifiable specs.

ChatDev 2.0 (DevAll) is an open‑source, zero‑code platform that lets you design and run multi‑LLM‑agent workflows via a visual web console or a tiny Python SDK. Define agents and their connections in YAML, launch from the browser or code, and use built‑in templates for data viz, 3‑D generation, game dev, research, etc. The backend is FastAPI, the frontend Vue 3, and the project ships Docker/Makefile support, extensible Python tools, and research‑grade orchestrators (Puppeteer, MacNet).

Aurelio is a self‑hosted wealth‑management app that stores all financial records locally and offers an optional AI chat advisor via OpenRouter. It aggregates accounts, portfolios, cash‑flow and goals, and the AI can answer queries, search the web, and suggest changes as confirmable cards.

An MCP server that provides AI coding agents with a curated library of real production websites and design tokens to improve the visual quality of generated UIs.

Foxel is a self-hosted, extensible cloud storage platform that unifies multiple storage backends and features AI-powered semantic search and an integrated AI agent for file management.

A comprehensive comparison platform that tracks and analyzes pricing, token quotas, and model availability across dozens of AI coding subscription services to help users find the best value.

A visualizer for neural network and deep learning models that supports a wide variety of model formats including ONNX, TensorFlow, and PyTorch.

An agentic discovery engine that helps AI agents find and rank the most suitable skills for a task by analyzing real-world execution performance and user feedback instead of just popularity.

Amagine3D is an open‑source web app that uses an LLM‑driven “3‑D‑native Agent” to turn product descriptions, images and dimensions into fully editable CAD models (build123d Python source) and manufacturing files (STEP, STL, 3MF). It validates wall thickness, clearances, and motion paths, supports multi‑part enclosures with hinges, and lets users tweak parameters in the UI. The stack combines React/Vite, an Express server, OpenAI‑compatible Codex, and Python CAD kernels (build123d + Open CASCADE).

A template vault for Obsidian that adds persistent, searchable memory to AI coding agents (Claude Code, Codex CLI, Gemini CLI) via hooks, slash commands, and optional local semantic search (QMD).

Open Terminal is a self‑hosted service that exposes a remote shell, file browser and code‑execution via a REST API, letting AI assistants run commands safely. It can be run in Docker (sandboxed with many pre‑installed tools) or installed as a Python package for direct host execution, supports configurable images, runtime package installation, optional multi‑user workspaces, and integrates with Open WebUI.

A demonstration of Claude Opus 5.5's capabilities, featuring three complex 3D web games generated entirely from one-shot prompts with zero human code intervention.

Atlas is a desktop IDE that adds AI‑agent‑aware source control. Every commit is linked to the exact agent session (prompt, tool calls, reasoning) via a local checkpoint database. It supports multiple agents (Claude Code, Codex, native Atlas agent, any ACP‑registered agent), shared on‑device memory, `@`‑mentions for files/notes/papers, a markdown knowledge base, Git UI, terminal, browser tab and optional team sync. All data stay on the machine by default; the app is distributed for macOS and Windows and can be built from source with Rust and Bun.

EasyWAM is a unified, high-performance research framework for training and evaluating World Action Models, designed to accelerate development through modular design and optimized training/inference pipelines.

A collection of high-quality SKILL.md files for AI agents (Cursor, Claude Code, etc.) that enable automatic task triggering for academic writing, diagramming, and development workflows.

Baserow is an open-source, no-code database platform that allows users to build applications and automate workflows using a spreadsheet-like interface, featuring an AI assistant for natural language development.

A state-driven AI persona engine for Telegram that simulates human behavior through sleep schedules, relationship stages, and long-term memory to avoid typical chatbot patterns.

OpenClaude Desktop is an open-source 1:1 recreation of Claude Desktop that runs on a Mac and lets you connect DeepSeek, Codex, Gemini, or any OpenAI-compatible AI — giving you the exact Claude experience without a Claude account or Anthropic subscription.

A curated library of 550+ medical research agent skills for Claude Code, Codex, and other agent platforms to support biomedical researchers across the entire research lifecycle.

An agent skill that helps AI write more natural, professional Japanese by enforcing writing principles and using morphological analysis to detect and remove common AI-generated cliches.

A collection of 65 expert-level frameworks and guided workflows for AI agents, helping users apply proven business, design, and coding principles from bestselling books to their projects.

A suite of MCP-based AI tools for Amazon sellers to automate competitor analysis, product selection, and listing optimization using data from multiple e-commerce services.

A Rust-powered Python client for the TongdaXing protocol, providing high-performance access to A-share real-time quotes, K-lines, and company data.

A comprehensive LLM evaluation framework that combines capability benchmarking, inference performance stress testing, and interactive result visualization in one tool.

Mimi is an open‑source, cross‑platform app that captures computer or microphone audio, sends it to speech‑to‑text and translation services, and shows the result as a floating subtitle overlay. It works on macOS, Windows, Linux and Android, lets you pick audio sources, adjust subtitle appearance, and optionally save subtitles or recordings.

A Python library for financial econometrics that provides tools for modeling volatility, testing for unit roots and cointegration, and performing statistical bootstrapping.

An open-source desktop content creation tool that automates the workflow of generating, editing, humanizing, and publishing articles for self-media authors.

SuperSplat is a free, open-source, browser-based editor for 3D Gaussian Splats. It lets you inspect, edit, optimize, and publish splat-based 3D scenes without installing any software.

OpenEdit is an open‑source CLI skill that lets AI coding agents (Claude Code, Codex, Gemini) edit videos by describing the desired changes. It handles transcription, layout, rendering in a headless browser and outputs an MP4, supporting subtitles, cuts, motion graphics and AI‑generated media.

A toolbox for experimenting with WebMCP, an experimental browser API that lets AI agents call structured tools on web pages. Includes a Chrome inspector, CLI evaluator, skill library, polyfill, and many demo apps showing both declarative and imperative tool usage.

Cloudflare Agents is a TypeScript SDK that turns Cloudflare Durable Objects into persistent, stateful “agents” with built‑in real‑time sync, scheduling, AI chat, tool execution, voice, email, payments, and more. It provides decorators for RPC, React hooks for the frontend, and a suite of packages for chat, code‑generation, and dynamic Workers. Ideal for per‑user assistants, collaborative apps, voice bots, and workflow engines.

A collection of open-source skills that enable AI agents to perform B2B prospecting, data enrichment, and lead scoring using real-time company data.

rayskills is a toolkit of 8 specialized AI skills for agents like Claude Code that automates professional workflows such as content production, knowledge management, and publishing while enforcing safety boundaries.

AiCode is an Android-based AI coding agent that provides a complete Linux development environment, allowing users to code, debug, and manage projects locally or via SSH.

An AI development framework that transforms natural language ideas into complete software projects by automating requirements gathering, task planning, and multi-agent execution.

An open-source communication infrastructure that provides a unified API and conversation model to connect products and AI agents to multiple messaging channels like Email, SMS, and Slack.

Agent Office is a TypeScript‑based tool that visualises LLM‑powered coding agents as desks in a 3‑D office. Each GitHub repo becomes a floor; agents can be spoken to, run terminals, manage git worktrees, and open PRs. The UI runs locally or on a cloud VM accessed via SSH tunnel, with deployment scripts for AWS, Azure, Railway, Fly.io, Dokploy and more.

An AI-powered creative workspace that uses agents to manage continuous visual production workflows, including image and video generation, on an infinite canvas.

transcribe.cpp is a C/C++ library for fast, cross‑platform speech‑to‑text inference. It supports 16 model families (60+ variants) via the GGUF format, runs on Metal, Vulkan, CUDA, or ROCm GPUs, and includes a tinyBLAS‑accelerated CPU path. The repo provides a model catalog with verified accuracy, a quantisation tool, official Python/TS/Rust/Swift bindings, and a CLI for easy transcription.

YOLOv5 is a PyTorch‑based, open‑source object detection/segmentation/classification model suite from Ultralytics, offering multiple model sizes, simple Python/CLI inference, one‑line training, and export to many deployment formats.

An AI-powered high-frequency trading bot for the Monad blockchain that uses the Jev model to capture the spread on the MON-USDC pair.

An open‑source tool that lets you describe a product in one sentence and, using LLM agents, automatically researches, designs style frames, captures 4K screenshots, generates music/voice (ElevenLabs optional), assembles scenes with Remotion, validates timing and audio, and outputs a finished promotional video.

AgentTeams is an open‑source platform that orchestrates multiple LLM‑driven agents (Workers) via a central Manager, using a Matrix‑based chat room for full human visibility and intervention. It bundles a gateway, shared file storage, and a Helm chart for Kubernetes, enabling secure, self‑hosted multi‑agent collaboration.

Apache Arrow is a universal, high-performance framework for in-memory data processing that allows different systems and programming languages to share and analyze large datasets efficiently without costly data conversion.

A repo hosting a three-part Microsoft Copilot Studio agent-building course, with contribution guides for different mission tracks and standard Microsoft open-source legal requirements.

A multi-engine AI coding desktop client that provides a graphical interface for various command-line AI tools, enabling users to chat, code, and manage Git operations without relying solely on a terminal.