AI applications

Local-first personal work-context desktop app for knowledge workers, unifying multi-source data to provide retrieval, Q&A, and digital-twin assistance for daily office tasks.

Self-evolving AI scientist system for exploratory vibe research, autonomously generating hypotheses, conducting experiments, producing insights, and iteratively improving its research strategies over time.

An open-source Suno alternative UI for ACE-Step that generates full AI songs locally with a Spotify-like experience.

Local-first, model-agnostic open-source workbench for scientific research that combines AI agents, code execution, and data connectors to produce reproducible and shareable reports and analyses.

Generates in-depth research reports by combining multiple large language models with live web retrieval, aimed at researchers and developers needing fast SSE API or MCP-server access.

An AI app that turns paper methods into editable SVG figures with style transfer and an in-browser vector editor.

A vision-based document intelligence service that converts complex PDFs, slides, Word files, and images into structured, RAG-ready chunks using OCR, layout analysis, and semantic segmentation.

AI research application that automatically decomposes static anime illustrations into multi-layer semantic layers, exporting PSD files for animators to use in subsequent animation and rigging.

Transforms unstructured text into structured graphs, hypergraphs, and spatio-temporal knowledge using LLMs, offering extraction, querying, and visualization through a single-command CLI.

Self-hosted job-hunting assistant for applicants applying DevOps-style tracking, aggregating listings from multiple platforms while scoring fit, refining resumes, and monitoring application progress.

An ultra-minimalist macOS recording and transcription app for meeting participants, capturing dual-track local audio and producing speaker-separated notes for quick review and follow-up documentation.

A free local stem-splitting app for musicians and producers that separates songs into vocals, drums, bass, piano, guitar, and other stems for practice, transcription, remixing, and export.

Unofficial Linux desktop client built from OpenAI's official macOS app, packaging Chat, Work, and Codex features with Debian, RPM, Arch, Nix, and AppImage installers for Wayland and X11.

Delivers an AI-powered English listening and speaking trainer using blind listening, intensive listening, shadowing, retelling, and spaced repetition to help learners master each audio lesson.

Real-time portrait animation framework for live streaming that generates expressive, arbitrarily long talking-head video from a single source image with natural motion.

A lightweight Windows app that pairs Live Captions with translation APIs for real-time speech translation.

An open-source content engineering system for GEO workflows that manages AI writing tasks, RAG-based semantic chunking, analytics, and multi-site publishing, including automated distribution to WordPress targets.

Open-source AI voice input app for macOS and Windows where holding a hotkey, speaking, and releasing inserts polished transcribed text at the cursor in any application.

A self-hosted open-source CRM and sales operating system with native AI agents and WhatsApp integration via WAHA, designed for chat-driven selling with multi-tenancy and MCP support.

AI-powered academic research application for fast literature search, one-click reviews, and full-text PDF translation, backed by a large built-in scholarly knowledge base with Zotero support.

A fast GPU-accelerated waifu2x image upscaler and denoiser shipping as portable executables for Windows, Linux, and macOS.

online image editor

Official TestSprite terminal tool for AI-powered automated testing, validating agent-built applications in real environments and returning detailed reports to developers.

An open-source AI interview platform built with Spring Boot, Spring AI, React, PostgreSQL/pgvector, and Redis. It supports resume analysis, mock interviews, voice interviews, and RAG knowledge bases.

A cross-platform CLI tool that brings AI chatbots from multiple free providers directly into the terminal.

A Windows AI upscaler app that enhances, upscales, and denoises photos and videos with an elegant easy-to-use GUI.

Research assistant plugin deeply integrated with Zotero libraries, enabling in-text Q&A, comparison, traceable citations, notes, and agent workflows for academic researchers.

Rust-built, AI-driven documentation engine that analyzes source code to generate structured repository wikis and agent-ready context for human teams and coding assistants.

Industrial-grade open-source framework for AI digital humans supporting real-time conversation, WebRTC playback, private deployment, and pluggable models for developers building interactive avatars.

A self-hosted X bookmark manager that turns saved tweets into a searchable, categorized visual knowledge base with AI-powered search.

A federated enterprise AI search application that queries 100+ business apps in place, delivering instant answers with citations while keeping sensitive data within existing security boundaries and deployments.

A blueprint-driven AutoResearch runtime that orchestrates AI research workflows from idea generation and experiments through paper writing and peer review with auditable evidence.

An AI-assisted tender document application for bidding teams, providing draft generation, knowledge base management, duplication checks, and compliance inspection in one workflow.

Lightweight locally runnable inference demo that launches chat, vision, and tool-calling services with one command for interactive testing by developers and users.

An AI-powered README generator that analyzes your codebase and produces polished, customizable documentation in one command.

In-browser Postgres sandbox with AI assistance (formerly postgres.new)

An open-source meeting transcription stack with auto-joining bots for Google Meet, Teams, and Zoom. It provides speaker-attributed transcripts, real-time WebSocket output, an API, and an MCP server.

Gives Mac users a free, fast open-source voice dictation app with quick transcription plus voice-driven text editing and rewriting, positioned as a lightweight alternative to Wispr Flow.

Biomes is an open source sandbox MMORPG built for the web using web technologies such as Next.js, Typescript, React and WebAssembly.

A high-performance Zotero AI and MCP plugin supporting many LLM providers. It enables cross-document synthesis over large PDF libraries with fast global search and traceable citations.

A long-form book translation workbench for translators and publishers that preserves terminology consistency across full manuscripts while supporting bilingual comparison, review, and proofreading of literary works.

A local-first AI e-book reader with RAG chat, semantic search, highlights, TTS, and cross-device sync.

Free, local-first open-source meeting and interview assistant offering real-time transcription, AI notes, lecture recording, local RAG, and stealth mode. Privacy-focused alternative with BYOK support and no subscription.

An automated pipeline that fetches trending Hacker News articles daily, generates Chinese summaries with AI models, and converts them into podcast audio using text-to-speech for Chinese-speaking listeners.

Web client prototype for scrcpy.

World-model application generating infinite interactive video worlds in real time on a single desktop GPU, supporting immersive exploration and interactive rollout experiments.

A lightweight Android browser with modern navigation

AI-powered news digest application that aggregates Twitter and RSS feeds into structured daily and weekly summaries managed through a web dashboard.

Shared spatial-thinking canvas combining handwriting, equations, diagrams, and visual reasoning, supporting collaboration with built-in or external AI agents for creators and teams.

A local-first wardrobe manager that extracts clothing items from photos, generates model try-on previews, and maintains a searchable on-device clothing database for users.

A desktop Price Action assistant that analyzes K-line data with two-stage AI diagnosis for discretionary traders.

Open-source desktop app for generating and editing videos with LTX models, supporting text-to-video and image-to-video workflows running locally or in the cloud.

Multi-agent team collaboration workspace built on OpenClaw, providing channels, documents, shared memory, and auditing. Supports coordinated work among AI agents in one environment.

An open-source, local-first conversational video editor where AI agents edit a real multitrack timeline.

A ComfyUI custom-node director for MiniMax H3 that orchestrates multi-segment text-to-video generation with audio, editing, and refining.

Self-hosted WeChat role-play chat service connecting to large language models, supporting multi-bot management, direct messaging, plus sticker and voice interactions.

Handwriting journal app for reMarkable Paper Pro that recognizes pen input and replies with LLM-generated responses rendered in flowing handwritten script for daily reflection.

Builds an open-source AI film-generation pipeline that orchestrates multiple video and audio models to produce longer, narratively coherent videos in a Netflix-style production workflow.

AI-native, self-hosted project management alternative to Jira and Trello where humans and AI agents collaborate on shared boards and sprints. Configurable through settings and plugins for Scrum teams.

Restores mosaic-censored JAV video and images using GPU acceleration and multiple detection-and-repair models, supporting video files, still images, and streaming processing workflows.

An open-source conversational AI customer-relationship system positioned as a Salesforce alternative, letting sales teams manage contacts, query knowledge bases, and execute routine tasks entirely through chat.

Local open-source AI short-drama and comic-drama generator covering story, storyboard, and video synthesis in one offline workflow, keeping user data on the local machine for creators.

A mobile app for AI-assisted comic-drama production that generates scripts, storyboards, and synthesized video in one flow. Enables creators to produce short dramatized videos quickly entirely on their phones.

Local-first AI chat application and inference engine for running open-weight LLMs privately and fully offline across devices, with OpenAI-compatible endpoints for agent use.

Daily AI-curated intelligence newsletter for indie hackers and builders that aggregates signals from ten-plus sources to surface one actionable startup direction with supporting evidence.

Adds plugin-based multi-agent support to in-editor academic writing, offering chat, revision, annotation, and critical literature-review assistance for researchers working inside Overleaf-compatible environments.

All-in-one AI short-drama and video creation platform for creators, orchestrating multiple models and skills across scriptwriting, storyboarding, and synthesis workflows.

Desktop macOS application for individual users to run and manage large language models locally, offering conversational chat plus API-compatible services on Apple hardware.

Open-source story engine for long-form fiction, giving authors persistent world memory and navigation assistance that tracks plotlines and advances drafts automatically.

A fast cross-platform desktop translator that captures on-screen text anywhere and returns accurate results from multiple translation engines.

macOS-native local-model workbench for Apple Silicon, integrating chat, image generation, MLX model management, serving, and performance monitoring in one desktop app.

A self-hosted personal internet radio station where an AI DJ curates music, hosts shows, and talks between tracks.

An open-source macOS AI copilot that floats over the screen, combining screen context with meeting audio to provide real-time answers while staying hidden from screen shares.

An AI-powered Word add-in that brings chat, translation, polishing, and agentic document editing right inside Microsoft Word.

Privacy-first AI notepad processes meeting recordings, transcription, and summarization locally on Windows and macOS, serving government, healthcare, legal, defense, and executive users handling confidential conversations.

Unifies video generation and editing by pairing an MLLM-based semantic planner with a DiT-based renderer, enabling high-quality synthesis and precise edits for video researchers and creators.

An AI-powered short comic-drama production platform that converts novel text into animated video, guiding creators through script analysis, character customization, storyboarding, and final synthesis.

A real-time speech translation desktop app that captures system audio, transcribes it, and shows translations in a minimal overlay.

A privacy communication app using local LLMs to hide encrypted messages inside natural-looking conversations, enabling covert secure dialogue across arbitrary everyday chat applications.

Desktop AI co-scientist that assists researchers across disciplines with autonomous data analysis, literature retrieval, chart generation, and report writing. Brings collaborative scientific workflows to local machines without specialized infrastructure.

Offline speech training system converting real-time voice to text and combining a 27,000-word emotion lexicon with AI analysis to diagnose expression issues and generate improvement reports.

Production-grade C++ edge AI engine for video analytics with visual pipeline orchestration, on-device VLM inference, real-time alerts, and on-screen display overlay.

An ultra-minimalist macOS push-to-talk dictation utility that transcribes speech while a key is held and inserts the resulting text at the cursor upon release for fast hands-free input.

Real-time Android game translation app for language learners and casual gamers, offering screenshot translation, floating word lookup, offline OCR, and speech narration during gameplay.

Tool that converts live website browsing traffic into usable APIs, supporting manual and agent-driven browsing while generating typed client code for developers.

Builds an AI-first marketing operations platform covering content strategy, multimodal generation, publishing, analytics, and SEO with connected management of social accounts for marketing teams.

Local-first desktop AI research workbench for scientific computing with Python and R, integrating literature, databases, MCP bioinformatics tools, GPU runtimes, and major LLM providers.

Local visual document parser that converts diverse files into AI-ready, searchable Markdown, supporting batch CLI processing and configurable models for knowledge workflows.

Supplies custom ComfyUI nodes and example workflows that call MiniMax H3 to generate speech and sound effects within visual creation pipelines.

An OpenCV-powered object speed camera for Raspberry Pi and PCs that tracks motion, measures vehicle speed, and serves images, data, and reports.

A local AI job-hunting app that collects postings, scores them with AI, and manages confirmed outreach and tailored resumes.

AIPPT Online editor，Base On ChatPPT， supports document editing services throughout the entire process, including import, export, layout beautification, online editing, playback, and presentation animation. It can quickly integrate into your products or internal systems of the enterprise.AIPPT在线编辑器，基于ChatPPT，支持整个过程的文档编辑服务，包括导入、导出、布局美化、在线编辑、播放和演示动画。

Fully local privacy-focused cross-platform speech-to-text app delivering multilingual transcription with neural models and on-device LLM post-processing for users needing offline workflows.

Supports A-share equity research with resilient workflows and evidence-grounded retrieval, producing versioned, reviewable reports alongside automated quality evaluation for analysts.

Local-first privacy guard that anonymizes documents before sending them to frontier LLMs and restores original text locally, supporting GDPR-compliant sharing workflows.

Local-first macOS global AI input completion app for Mac users, running on-device models to provide inline suggestions, emoji macros, autocorrection, and text expansion system-wide.

An AI-powered multi-voice audiobook generator built on Qwen3-TTS, combining LLM script annotation, voice cloning and design, per-line style control, and MP3 or M4B export.

Free open-source invisible overlay assistant for technical interviews, coding assessments, and algorithm practice, providing real-time AI help, screenshot question capture, and multi-language support.

Delivers an open-source, AI-native SOC platform for alert triage, ticket collaboration, correlation detection, and automated orchestration of security response workflows for analyst teams.

Performs street-level geographic localization from a single image through three-stage retrieval and matching, reporting precise street coordinates for geolocation analysts.