Trending AI projects

OpenBB’s Open Data Platform is an open‑source Python library that unifies financial data access, offering a pip‑installable API, a FastAPI server, and CLI, and can be linked to the commercial OpenBB Workspace for analyst dashboards and AI‑agent integration.

Zephium is an open-source Rust browser using your OS's native web engine (WebKit on macOS, WebView2 on Windows). It blocks ads/trackers by default, runs Chrome extensions, and includes tasks, notes, and time tracking. Its standout feature is Work, a visible canvas where AI agents carry out tasks in parallel while you watch; Ask Zephium is a sidebar assistant. It's private by default (no telemetry, no account needed), supports macOS Apple Silicon and Windows x64, and is licensed under MPL-2.0.

It's a Plan is an open-source, self-hosted project management and issue tracking platform where AI agents work as full team members with roles and permissions, sharing the same boards as humans. It includes standard project features like Kanban, cycles, and dashboards, and exposes everything via REST API, webhooks, and MCP.

Microduck RL provides MuJoCo‑Warp environments and PPO training scripts for the 800 g bipedal Microduck robot. Policies (walk, stand, tricks, etc.) are trained with realistic BAM actuator models, exported to ONNX, and can be published to Hugging Face for instant deployment on the real robot via the companion runtime.

cmux is a native macOS terminal built on Ghostty that adds vertical tabs, notification rings, an in‑app browser, SSH workspaces, and a programmable CLI to help developers monitor and orchestrate multiple AI coding agents such as Claude Code, Codex, and OpenCode.

BMad Method (BMAD-METHOD) is an open-source, AI-driven development methodology that helps you turn ideas or change requests into working software. It keeps decisions explicit, carries context forward, and sizes its process to the work, so a small change goes straight to build while complex projects get deeper planning.

Agnes Harness is a developer-focused framework for building and deploying AI agents that integrate with enterprise business systems and physical hardware, emphasizing auditability, human-in-the-loop approvals, and reusable plugin architectures.

A Model Context Protocol (MCP) server that enables LLMs to automate web browsers using structured accessibility snapshots instead of screenshots.

An open-source chat interface that turns AI-generated text answers into interactive tools, 3D visualizations, and dynamic UI components.

A comprehensive kit of instructions, design tokens, and automated verification gates that turns AI agents into senior UX/UI architects for producing accessible, production-ready interfaces.

A comprehensive, daily-updated directory of over 500 free LLM APIs that helps developers find, compare, and integrate free models into their existing tools using standard OpenAI-compatible endpoints.

AnyMove is a motion capture tool that converts AI-generated two-view videos into 3D animations for rigged creatures, allowing users to animate complex models without manual keyframing.

dsh-image-gen is a plugin for DeepSeek Harness that brings a complete AI image creation workflow into the platform, enabling chat-based generation/editing, an infinite canvas, batch studio, multi-model comparison, a 500+ prompt inspiration library, and local ComfyUI integration with support for multiple cloud providers and subscription-based login without API keys.

A self-hosted, open-source AI chat platform that unifies multiple AI providers into a single, privacy-focused interface with support for agents, code execution, and multi-user management.

A specialized skill for AI coding agents that refines AI-generated Japanese text for better readability and structural clarity in technical and business contexts, while providing tools to verify that the meaning remains unchanged.

An AI‑driven toolkit that lets LLM agents (Codex, Claude, Cursor, etc.) control real SolidWorks (and other CAD tools) via natural language. It offers three interchangeable entry points—Skill, MCP client, and a Windows CAD Studio UI—all sharing a common capability list and an execution core that runs COM/OCCT commands, verifies results, and records evidence. Verified capabilities include part modeling, assembly, drawing, export, and more; unverified ones are clearly marked. The project emphasizes verification‑first success, deterministic reliability testing, and easy installation on Windows.

Claude Trading Skills is an open‑source toolkit of Anthropic Claude Skills that helps individual investors run structured, human‑in‑the‑loop trading workflows (market review, risk management, trade planning, journaling, and post‑trade analysis). It provides modular skill definitions, YAML workflow manifests, and ready‑to‑upload .skill packages for use with Claude Web or Claude Code.

DSH Pocket is a plugin for DeepSeek Harness that enables remote mobile access to your local AI agent interface via secure LAN or public internet tunnels.

An auto-evolution framework that uses AI agents to iteratively optimize the prompts, configurations, and code of AI agent systems and workflows based on failure analysis.

NullMoth NVIDIA driver is an open-source Metal driver that lets Turing-and-newer GeForce cards work on macOS Sequoia and Tahoe. It translates Metal shaders to Vulkan via a custom pipeline, with kernel extensions running NVIDIA's own GPU modules, enabling Hackintosh and Intel Mac users to use these GPUs.

Open Steps is a MIT‑licensed pack of prompt‑templates (“skills”) and two shell hooks that make Claude Code, Codex CLI, Cursor CLI and Gemini CLI speak plain language. It provides one‑screen “done‑or‑not” reports, step‑by‑step user instructions, simple question/answer prompts, premortem risk analysis, backlog tracking, and even autonomous PR merges. Install via a Claude plugin or by copying the skill folders and adding a tiny routing block to each agent’s config.

Drama Skills is an open‑source, agent‑compatible toolkit that lets writers turn a novel or idea into a finished short‑drama video. It provides eleven “skills” (commands) that cover the whole pipeline—analysis, script writing, asset extraction, image/video prompt generation, storyboard, production, editing, and review—using plain Markdown files as the single source of truth. Install via ClawHub, run the commands inside Claude Code, Codex, or any Agent‑Skill environment, confirm media generation before spending, and export a complete, auditable project.

TencentDB Agent Memory is an open‑source memory hub for LLM agents. It captures conversations, extracts reusable skills, builds a Wiki from docs, and indexes code into a CodeGraph. A central UI (Memory Hub) lets teams manage assets, set fine‑grained ACLs, and bind specific memory pieces to individual agents. Agents talk to a uniform proxy, so any framework (Claude Code, DeepSeek Harness, OpenClaw, etc.) can use the same memory without code changes. Benchmarks show a 59 % boost in persona retention. The repo provides one‑click Docker deployment, migration tools, and a roadmap for zero‑config cold starts and more integrations.

Qingjian is a cross‑platform Chinese input method that shows a one‑word translation (English, Japanese or Spanish) next to each candidate and uses a small local model to suggest whole‑sentence completions. All processing stays on the device, with optional user‑provided cloud completion. It runs on macOS, Windows and Linux (Fcitx), tracks typing statistics, and is released under GPL‑3.0‑or‑later.

PriceAI is a neutral price-comparison radar that aggregates data from over 200 AI subscription channels, official services, and API transit providers to help users make informed, cost-effective purchasing decisions.

AX is a Kubernetes‑style orchestrator for autonomous AI agents. It lets you declare a Workspace (code and tools), a Task (the agent execution), and a Model (LLM config) in YAML, then runs the agent in a sandboxed container on top of Agent Substrate. Features include high‑throughput scheduling, resource limits, suspend/resume, and interactive `ax ssh` debugging.

A code repository accompanying a book that teaches you to build a reasoning LLM from scratch, starting from a pre-trained model and adding reasoning capabilities via inference-time scaling, reinforcement learning, and distillation — all runnable on consumer hardware.

An all-in-one AI academic writing platform that assists with drafting, editing, formatting, and presentation creation, while offering white-label solutions for businesses to launch their own branded writing services.

AiToEarn is an AI-powered platform that automates the entire content marketing lifecycle, from AI-assisted creation and multi-platform distribution to community engagement and monetization.

A Python library and MCP server that provides programmatic access to Google Flights data via reverse-engineered APIs, avoiding web scraping.

Cline is an open-source autonomous coding agent that integrates into your IDE, terminal, and desktop to perform complex development tasks, execute shell commands, and manage multi-agent workflows.

The Hugging Face Agents Course is a free online course that teaches building and evaluating AI agents, covering frameworks like smolagents, LangGraph, and LlamaIndex, plus agentic RAG and fine-tuning for function calling.

A phone/browser client for herdr that turns AI coding agent sessions into chat conversations with live-terminal access, tap-to-approve prompts, file viewing, and push alerts.

A universal account management tool for AI IDEs that enables one-click account switching, quota monitoring, and running multiple isolated instances of AI coding tools.

awesome-agentic-ai-zh is a free, MIT-licensed learning roadmap (in Traditional Chinese, with Simplified Chinese and English versions) that guides readers from AI Agent basics to building reliable agent systems. It offers 10 learning stations (Stages 0–8 plus 7.5), two main tracks (CLI Power User for using existing agents, Agent Builder for building your own), role-based branches for different professions, runnable exercises, plain-language explanations of technical terms, and time estimates. It's a curated path that points to external resources rather than being an encyclopedia.

RAG-Anything is an all-in-one multimodal RAG framework that enables seamless processing and querying of complex documents containing text, images, tables, and mathematical equations.

jevgrep is a Node‑based CLI that lets AI coding agents locate relevant source files and snippets by asking natural‑language questions about a repository. It sends filtered code to a Jev‑compatible LLM endpoint, returns concise file lists and code excerpts, and has been shown to cut coding‑agent inference cost by about 30 % while maintaining the same task‑success rate.

An open-source AI Trading OS that enables users to turn AI-driven market research into Python strategies, backtests, and live execution across various brokers and exchanges.

Motion2MixamoRig is an AI-powered tool that extracts human motion from videos or photos and retargets it onto 3D characters rigged for Adobe Mixamo, producing ready-to-use animation files for game development.

A general-purpose problem-solving engine that uses a state-space search and blackboard architecture to autonomously find paths to goals, specifically validated for AI penetration testing.

OpenWiki is a Node‑based tool that lets LLM coding agents automatically generate, query, and keep up‑to‑date a linked Markdown wiki inside a code repository. It integrates with many agents, supports parallel page generation, workspace linking, visual graph browsing, and CI‑driven updates, all while storing plain Markdown you own.

SillyTavern is a locally installed user interface that provides a unified frontend for interacting with various LLM APIs, image generation engines, and TTS models.

A blazingly fast 3D animation engine powered by Rust and wgpu that is compatible with the Manim CE API and optimized for AI agents to generate animations.

RustFS is a high‑performance, distributed object storage system written in Rust. It provides full S3‑compatible APIs (versioning, encryption, lifecycle, replication, etc.), native OpenStack Swift support, a web console, and optional protocols like FTPS/WebDAV/SFTP. Distributed deployments can run on Docker, Kubernetes (Helm chart), or as a single‑node binary, and it integrates with Vault/AWS KMS, OIDC, and webhook notifications. Licensed under Apache 2.0, RustFS is positioned for AI/ML data‑lake workloads, edge devices, and enterprise‑grade storage.

Aliens Eye is an AI-powered OSINT username scanner that checks 840+ platforms quickly, using a blend of a trained machine-learning model and structural page features to classify each hit as Found, Maybe, or Not Found with reduced false positives. It includes cross-site correlation, recursive expansion, domain checks, watch mode, and various output formats for investigators and security researchers.

A scientific PDF translation tool that preserves original layouts, formulas, and charts while supporting multiple translation services and OCR.

An AI-powered McDonald's fan certification tool that uses official MCP interfaces to generate personalized meme cards, suggest meals based on real nutritional data, and facilitate coupon claiming.

LangExtract is a Python library for LLM‑driven, grounded extraction of structured data from any text. It chunks long documents, runs parallel LLM calls (Gemini, OpenAI, Ollama or custom providers), maps each extraction back to its exact character span, and produces an interactive HTML visualiser. Install via pip, define a prompt + few‑shot examples, call `lx.extract`, and save/visualise the results. Supports cloud, local, and batch processing, with a plugin system for new model providers.