Trending AI projects

An intelligent customer service agent for robot vacuum users that combines Agentic RAG and device data analysis to provide technical support and personalized usage reports.

A Python API and database of CAN bus definitions that allows developers to read vehicle data and send control commands to modern cars for advanced driver-assistance applications.

A set of plugins and style guides that enable AI agents to communicate using idiomatic Russian profanity to provide more concise and emotionally honest technical summaries for developers.

An agent orchestration system that splits complex tasks into specialized roles and staffs them with verified domain experts from local or public sources to minimize token usage.

An end-to-end solution for creating digital avatars by fine-tuning LLMs on exported chat history, featuring data preprocessing, privacy filtering, and multimodal support.

A Claude Code skill that teaches the AI concrete, mechanical design rules from the book Refactoring UI to create professional-looking user interfaces.

Agent Storyboard is a local‑first Node.js web app that lets Codex or Claude Code agents generate full video storyboards—script, shot list, AI‑generated images, motion clips, and voice‑overs—directly into a single spreadsheet‑style table. All data and media stay on your computer; the agent interacts via MCP calls, not a browser. Features include drag‑and‑drop shot ordering, batch generation, built‑in visual styles, Whisper‑based voice alignment, and export to Markdown/HTML/Word.

A comprehensive, hands-on tutorial series for beginners to learn LLMs and AI agents through six progressive implementation tasks, from building a mini-GPT to creating a coding agent.

A local AI companion and interactive storytelling application that enables users to create persistent characters and worlds with evolving memories and relationship tracking.

An MCP server that enables AI assistants to control and interact with iOS simulators, allowing for automated UI testing, app navigation, and screen inspection.

Codex Switcher is a desktop utility that allows users to manage, monitor, and quickly switch between multiple OpenAI/ChatGPT accounts while tracking usage quotas and automating account warm-ups.

A library of highly optimized kernels for LLM training and inference, providing high-performance implementations of MoE routing, quantization, and Engram gating.

OpenAPPA is a deterministic, policy‑driven guardrail that sits between LLM agents and their tools, checking each tool call against a TOML‑based permissions policy to prevent unauthorized data flows. It offers a CLI for policy validation, integrations (e.g., Claude Code, Archestra proxy), and shows strong security benchmarks while maintaining high task‑completion rates.

A Video Agent that transforms long-form videos into structured, searchable knowledge with every AI-generated conclusion backed by verifiable timestamped evidence.

A macOS tool that enables AI agents to access and search live WeChat chat history locally, allowing users to turn conversations into structured personal notes.

An interactive scientific agent and workspace that helps researchers analyze papers and data using a recursive self-improvement framework to refine its analysis procedures and models.

CanIRun.ai is a browser-based tool that automatically detects your CPU, RAM, and GPU using standard web APIs, then recommends the best open-weight AI models to run locally for coding, chat, reasoning, and vision tasks. It grades each model S–F based on your hardware, with no installs or benchmarks needed.

RoboLab is a high-fidelity simulation benchmark built on NVIDIA Isaac Lab that provides over 100 manipulation tasks to evaluate and compare generalist robot policies.

A self-hosted, OpenAI-compatible LLM router that automates model selection based on cost and capability, providing a unified API for multiple AI providers.

A Python framework for agent-based modeling that allows users to create, visualize, and analyze simulations of complex systems.

A C++ plugin for Godot 4 that provides Behavior Trees and Hierarchical State Machines to help developers create complex, modular AI behaviors for game agents.

An AI-powered penetration testing tool that uses LLMs to automatically identify login form selectors and execute automated brute-force attacks.

A curated directory of open-source MCP servers, AI skills, and plugins that enable AI agents to interact with and automate SAP software and development workflows.

GitLab MCP Server is a lightweight, install‑once binary that exposes GitLab’s API as 275 MCP‑compatible tools (issues, merge‑requests, pipelines, wiki, releases, etc.) for AI agents. It supports PAT or OAuth authentication, multiple transports (stdio, SSE, streamable HTTP), fine‑grained permission modes, and can run locally, in Docker, or on custom GitLab instances. Designed for Claude, Codex, Cursor, VS Code, Copilot and other MCP clients, it lets LLM‑driven workflows manage GitLab resources without writing custom code.

A professional AI subtitle software that uses advanced ASR models and a custom Rust/C++ engine to provide high-accuracy local transcription and translation.

Magika is an open‑source, AI‑driven file‑type detection tool. A tiny deep‑learning model (few MB) trained on ~100 M files can label over 200 binary and text formats with ~99 % accuracy in ~5 ms per file on a single CPU. It ships as a Rust CLI and as bindings for Python, JavaScript/TypeScript, and (work‑in‑progress) Go. Install via pip, brew, cargo, or npm, then run `magika <path>` or use the Python `Magika` class to get structured results (label, MIME type, confidence score). Used internally at Google for security scanning and integrated with VirusTotal and abuse.ch.

OpenAI’s gpt‑oss repo delivers reference code and tooling for the open‑weight 120 B and 20 B LLMs, including a special Harmony chat format, MXFP4 quantization, and back‑ends for PyTorch, Triton, Metal, vLLM, Ollama, and more. Licensed Apache‑2.0, it’s aimed at researchers and developers who want to run, fine‑tune, or build tool‑augmented agents with these models.

A private, on-premise document intelligence platform that uses hybrid search and local LLMs to let users query and summarize their files without data leaving their machine.

A procedural generation engine that creates infinite photorealistic 3D worlds, indoor scenes, and articulated assets for computer vision and robotics simulation.

Paperless‑AI is an open‑source extension for the self‑hosted document manager Paperless‑ngx. It automatically classifies new documents, adds AI‑generated titles/tags, and builds a vector index so you can ask natural‑language questions across your whole archive (RAG chat). It works with any OpenAI‑compatible model—including local Ollama servers—and ships as a Docker image with a simple web UI for configuration and manual processing.

A simulated bash environment with a virtual filesystem that allows for safe execution of bash-like commands in a sandbox.

A mobile AI workspace for iOS and Android that lets users chat with multiple AI models, build custom agents with tools, and generate images.

TotalSegmentator is an open-source deep-learning tool that automatically segments dozens of anatomical structures (organs, bones, vessels, muscles, etc.) in CT and MR images, outputting 3D masks and statistics. It saves radiologists and researchers hours of manual annotation and supports many task-specific models for different body regions.

Lucebox is an open‑source C++ inference engine for large language models that uses speculative pre‑fill and custom GPU kernels (CUDA/HIP) to dramatically speed up token generation on consumer and mixed‑GPU setups. It supports many models (Qwen, DeepSeek, Gemma, etc.), offers vision input, and provides ready‑made hardware‑specific profiles for AMD and NVIDIA cards, delivering hundreds of tokens per second on a single GPU and even higher throughput in heterogeneous configurations.

A series of billion-parameter text-to-3D human motion generation models that create skeleton-based animations from text prompts using Diffusion Transformers and Flow Matching.

CoreCoder is a 2,800‑line pure‑Python coding‑assistant engine. It implements the full agent loop (LLM call → tool execution → context update) with eight built‑in tools, cost tracking, three‑tier context compaction, sub‑agents, and a REPL. Designed to be read in an afternoon and forked, it serves as a transparent reference for building or customizing AI coding agents.

FanBox is a local-first desktop cockpit for managing AI coding agents, providing a unified interface to browse files, run terminal commands, and track AI-driven code changes in real-time.

An environment and benchmark for building and testing autonomous agents that can control Android devices across a wide variety of real-world apps.

A collection of agent skills that automates the discovery, critical review, and structured note-taking of academic research papers from arXiv and Hugging Face.

A GitHub Action that integrates Claude Code into CI/CD workflows, enabling automated agentic coding tasks like PR reviews and dependency updates.

A browser-based studio for building, curating, captioning, and training image datasets for LoRAs, integrating ai-toolkit and ComfyUI.

Kite is an open-source Kubernetes workspace for multi-cluster operations, observability, and AI-assisted troubleshooting.

hyprwhspr is a native Linux speech‑to‑text app that runs as a system‑wide service. It uses local models (Whisper, Cohere, Parakeet, Qwen3‑ASR, etc.) for near‑real‑time dictation, optionally falls back to cloud APIs, and integrates with Wayland/X11 desktops, status bars, and a themed visualiser. Install via the AUR or a one‑line script, configure with `hyprwhspr setup`, then toggle dictation with `Super+Alt+D` to have transcribed text auto‑pasted into any window.

MoveIt 2 is an open-source motion planning framework for ROS 2 used to develop robotic manipulation applications and benchmark planning algorithms.

MrNeRF/awesome-3D-gaussian-splatting is a curated resource list for 3D Gaussian Splatting, organizing papers, implementations, viewers, tools, and learning materials into a single searchable directory.

A debugging tool for Claude Code that reconstructs session logs into a visual interface to inspect tool calls, token usage, and hidden reasoning steps.

An MCP server that enables LLMs to inspect MySQL schemas and execute SQL queries, featuring built-in PII redaction and secure remote access.

Bolna is an end-to-end open-source platform for building production-ready voice conversational agents. It orchestrates speech recognition, LLM reasoning, and speech synthesis over websockets, supporting multiple telephony, ASR, LLM, and TTS providers, and can be configured via JSON or a simple Python API.