A framework for building real-time, multi-modal voice agents that can see, hear, and understand, featuring integrated job scheduling and flexible model integrations.

Pinvou Agent is a cross‑platform desktop AI‑assistant that lets you work, design, or code with LLMs. It combines chat with file attachments, a local knowledge base, persona cards, and a plug‑in tool store, producing editable artifacts (Markdown, designs, code) that persist across sessions. Built with React + Vite UI on Tauri 2, it uses the CodeWhale engine for model calls, MCP tool execution and Skills. Supports local vLLM or any OpenAI‑compatible endpoint, runs on Windows/macOS/Linux, and is MIT‑licensed.

An open-source toolkit that automatically generates movie recap videos with AI-powered scripts, narration, subtitles, and rendered video from a single prompt.

Read Frog is an open‑source browser extension that uses LLMs and AI TTS to translate, explain, and turn web content into flashcards, offering bilingual view, context‑aware translation, subtitle translation, custom AI actions and support for 20+ AI providers.

ExcelMcp is a Windows‑only server and CLI that let LLM agents control a live Microsoft Excel instance via its COM API, exposing 326 operations (Power Query, PivotTables, VBA, Python, etc.) for real‑time spreadsheet automation.

A tool for AI agents to transform video frames into professional 3:4 social media quote images using either original burned-in subtitles or custom script-based overlays.

iMCP is a macOS app (with a bundled `imcp‑server` CLI) that implements the Model Context Protocol, exposing Calendar, Contacts, Location, Maps, Messages, Reminders and Weather as JSON‑LD tools for AI clients like Claude Desktop, Claude Code, Cursor and Amp. Users enable services via macOS permission dialogs, connect the AI client, and then can ask natural‑language questions that the AI fulfills by fetching real personal data locally. The project includes Swift SDK integration, a custom Swift package for reading the iMessage database, and debugging tools (MCP Inspector, Companion).

An open-source design editor that natively reads .fig files and provides a programmable toolkit for AI-driven design automation and custom editor development.

An end-to-end AI production workspace for short dramas that manages the full pipeline from script breakdown and asset consistency to video generation.

An open-source multimodal AI workbench that integrates image, video, and audio generation into a node-based infinite canvas for iterative visual creation.

Gas Town is an open‑source Go‑based workspace manager that lets you run many AI coding agents (Claude, Copilot, Codex, Gemini, etc.) on multiple Git projects while persisting every step in a Git‑backed ledger. A central AI coordinator (the *Mayor*) creates *convoys* of work items (*beads*), hands them to worker agents (*Polecats*), and a set of watchdog services (Witness, Deacon, Dogs) keep the system healthy. Completed work is merged via a Bors‑style queue (*Refinery*), and an optional federation layer (*Wasteland*) lets different installations share tasks. Installable via Homebrew, `go install`, or Docker, it provides a CLI (`gt`) for creating rigs, crews, convoys, and for monitoring progress.

An AI-powered tool that extracts individual garments from photos to create a digital wardrobe and generates modeled editorial previews of outfits.

An automated pipeline that transforms text prompts or images into interactive 3D holographic cards with depth and perspective-shifting shimmer for the web.

Mnemosyne is a local-first, SQLite-backed memory layer for AI agents. It gives assistants persistent memory across sessions via a three-tier architecture (working memory, episodic memory, temporal knowledge graph), with hybrid semantic/keyword search, heavy vector compression, a Python SDK, CLI, and built-in MCP server. It's privacy-focused: no telemetry, local by default, with optional client-side-encrypted sync.

ha‑mcp is an unofficial Home Assistant Model Context Protocol server that lets LLM assistants (Claude, ChatGPT, Gemini, etc.) fully read, control, and configure a Home Assistant instance. Install via the HA‑MCP custom component (HACS) or as a Home Assistant add‑on/Docker/PyPI server, obtain a secret webhook URL, and point your AI client at it. The server exposes 87 tools covering device control, automation/script creation, dashboard editing, backups, logs, and safety policies, enabling natural‑language management of the entire smart‑home setup.

An automated AI-powered tool for movie and TV commentary videos that handles scriptwriting, video editing, voiceovers, and subtitles in one workflow.

HelloAgents is a Python library that provides a production‑grade framework for building multi‑agent AI applications. It wraps OpenAI‑compatible, Anthropic, and Gemini LLMs, offers built‑in tools (file I/O, task delegation, todo tracking), and includes engineering features such as context management, session persistence, circuit‑breaker, optimistic locking, observability, streaming SSE, and async lifecycle. Install via `pip install hello-agents`, configure a `.env` with your API credentials, and start agents like `ReActAgent` with a registered tool set.

An open-weight 11B-parameter model series for real-time and offline video understanding, featuring a cross-attention architecture that supports interruptible, streaming interaction.

A real-time interactive streaming engine for digital humans that synchronizes audio and video for lifelike virtual avatars driven by LLMs and TTS.

An open-source video translation tool that automates speech recognition, subtitle translation, and AI dubbing with support for voice cloning and multi-role synthesis.

A cross-platform screen search tool that integrates screenshotting, offline OCR, translation, and reverse image search to extract and act on screen information.

An experimental generative UI operating system that replaces traditional apps with live-streamed, native mobile interfaces generated on the fly based on user requests.

Easydict is a native macOS app for fast word lookup, translation, and OCR. It auto‑detects language, offers several shortcut‑driven input modes, and aggregates results from 20+ services—including classic dictionaries, cloud translators, and LLMs like OpenAI, Gemini, Claude, and Ollama. Install via Homebrew (`brew install --cask easydict`) or a manual download. Open‑source (GPL‑3.0) and welcomes community contributions.

A library and CLI tool to remove visible labels, invisible pixel watermarks, and AI provenance metadata from AI-generated images and videos.

vibe‑motion/skills is a plug‑in library that lets AI agents (e.g., Claude Code) generate a variety of animated visual assets—ruler progress bars, fisheye video effects, brand launch clips, 3‑D Earth routes, vinyl player animations, etc.—by cloning and running dedicated rendering projects. Install via `npx skills add vibe-motion/skills`, pick a skill, provide parameters, and receive a GIF/MP4/HTML output.

LLPlayer is a Windows‑only C# media player that adds AI‑driven subtitle generation (Whisper ASR), real‑time translation (via cloud or local LLMs), OCR for bitmap subtitles, dual‑subtitle display, word lookup, and online video support. It targets language learners and is open‑source under GPL‑3.0.

An automated pipeline and desktop app that converts layered PSD files into fully rigged Live2D models, featuring adaptive meshing, automatic deformer setup, and AI agent integration.

A deployment recipe for serving the Qwen3.8-Flash-Next vision-language model on a single DGX Spark using vLLM, featuring memory-mapped PLE tables and speculative decoding.

A multi-agent framework that generates editable CAD parts and multi-part assemblies from natural language, featuring automatic code repair and simulation-ready exports.

Open‑AI‑Canvas (影策) is an open‑source, self‑hosted web workbench that lets creators turn a text brief into a complete set of film‑style assets—storyboards, character/style libraries, and AI‑generated images, video and audio—through an integrated canvas, async task queue and local AI‑agent system.

An AI-assisted workflow and toolset for converting raster logos into clean, animated SVG logos with a focus on precision fitting and HTML-based motion delivery.

BabelDOC is an open‑source Python tool that translates PDF scientific papers using LLMs (OpenAI‑compatible models) and outputs bilingual PDFs. It offers a CLI, a Python API, extensive PDF‑handling options, glossary support, offline asset packaging, and can be run locally or via a hosted beta service.

A comprehensive prompt engineering library for Higgsfield AI that transforms natural language into production-ready cinematic video and image prompts.

An open-source framework that uses coding agents to transform game requirements into production-ready assets and engine-specific code for UE5, Unity, Godot, Blender, and three.js.

BettaFish (微舆) is an open‑source, Python‑based multi‑agent platform that automatically crawls, analyses and reports on public opinion from dozens of social‑media sources. Users ask a question in a chat‑like UI; three specialized agents (Query, Media, Insight) gather data, a ForumEngine lets them debate under an LLM moderator, and a ReportEngine assembles an interactive HTML (or PDF/Markdown) report. The system includes a 24/7 crawler (MindSpider), multimodal video/image processing, fine‑tuned sentiment models, and supports private‑data integration. It ships with Docker compose, a Flask front‑end, Streamlit demos, and is extensible to other domains.

An open-source AI-native vector design tool that generates UI layouts from natural language prompts and exports them to production code.

A backend server for the xiaozhi-esp32 project that enables ESP32 devices to become AI assistants with voice, vision, and smart home control capabilities.

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).

RAG‑Anything is a Python library that builds a multimodal Retrieval‑Augmented Generation system. It parses PDFs, Office files, images, tables, and equations using MinerU, creates a cross‑modal knowledge graph, and retrieves relevant pieces with a hybrid vector‑graph search before passing them to an LLM for answer generation. Install via `pip install raganything`, configure with `RAGAnythingConfig`, ingest documents, and query in natural language.

A Python SDK that lets you build and consume servers following the Model Context Protocol, a standardized API for LLM‑to‑tool communication. It provides decorators for exposing functions as LLM‑usable tools/resources, supports stdio/HTTP/SSE transports, includes a CLI, and is documented at py.sdk.modelcontextprotocol.io.

4DAnyone is a research‑grade neural model that converts a single‑camera video of a person into dozens of synchronized, view‑consistent videos (6‑24‑48 views) for downstream 4‑D Gaussian Splatting reconstruction. It runs on consumer GPUs (≤24 GB VRAM), offers a fast distilled Turbo variant (5.6× speedup), and provides ready‑to‑use inference scripts and nerfstudio integration.

A native Metal inference engine for MiniMax-H3 on Apple Silicon, enabling local text-to-video and text-to-audio generation with advanced memory and performance optimizations.

A self-hosted file manipulation suite that provides over 200 tools for converting, compressing, and processing images, video, audio, and PDFs using local AI for privacy.

A lightweight pure C/C++ implementation for running image and video diffusion models locally, supporting a wide range of hardware backends and model architectures.

An all-in-one video subtitle tool that uses LLMs for speech recognition, subtitle optimization, and translation to automate the video captioning process.

SenseNova‑Skills is an open‑source collection of Agent‑Skills that add office‑automation functions (image generation, infographic creation, Excel analysis, deep research, and PowerPoint generation) to SenseNova LLMs. Drop the skill folders into an OpenClaw or hermes‑agent runtime, configure the SenseNova API, and the agent can turn a simple natural‑language request into polished data‑driven reports and decks.

Kimi K3 is an open-weight, 2.8T-parameter native multimodal MoE model designed for long-horizon coding, deep research, and complex agentic knowledge work.

An open-source diffusion model engine that enables high-performance inference and training of image, video, and audio generative models on consumer-grade GPUs.