PYTHIA is a fully local global-intelligence and forecasting app that fuses dozens of keyless live feeds into a 3D world view and uses a local LLM to predict what happens next.jangles-byte
A browser-based batch subtitle translator for SRT, ASS, VTT, and LRC files. It translates whole seasons into 120+ languages using traditional translation APIs or 27 LLM providers while keeping original timing intact.rockbenben
An open-source desktop AI copilot for technical interviews. It shows a stealth overlay invisible to screen sharing, takes voice or screenshot questions, and streams real-time answers from Gemini.techycsr
vision-demos is a collection of runnable computer-vision demos that use pose and segmentation models to analyze dance sync, pull-ups, climbing routes, basketball shots, running and deadlifts.jeremyipark
Vigolium is a vulnerability scanner that combines a deterministic native scanning engine with agentic AI for autonomous discovery, exploitation testing, source-code audit, and finding triage.vigolium
A private AI chat interface that runs large language models directly in your browser using WebGPU and WebLLM. It supports multiple models, vision input, and custom MLC endpoints without sending data to a server.mlc-ai
An AI mock-interview platform backend built on Spring AI, offering resume analysis, multi-agent interview scoring, AI chat, and speech transcription and synthesis.lishuangqiang
LaTeXSnipper is a desktop math OCR workspace that turns screenshots, images, PDFs, and handwriting into editable LaTeX, Markdown, and other formats. It bundles local recognition models, a live formula editor, Office plugins, and a scripting API.sakuramathcraft
Alexandria turns books into multi-voice audiobooks using an LLM for script annotation and Qwen3-TTS for voice cloning, voice design and LoRA training, exporting MP3, M4B or Audacity projects.finrandojin
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。YOOTeam
Rescript is a transcript-based video and audio editor that transcribes media locally with Whisper, letting you cut by deleting words and export the result in many formats.wassgha
ASReview LAB is an open-source active-learning application for systematic reviews and text screening. Its AI models learn from relevance decisions to prioritize records, with an interactive interface, simulation toolkit, and export options.asreview
A self-hosted dark-web OSINT and threat-intelligence platform that turns one query into a sourced report, searchable entity graph, and exportable indicators through an automated multi-step pipeline.KatrielMoses
macOS utility that divides the MacBook desktop into four tappable zones, locally classifying tap locations to trigger corresponding user-defined actions without sending data elsewhere.justingamer191
Vocalinux is an offline voice dictation desktop app for Linux that runs local speech-to-text engines like whisper.cpp and VOSK and types the result into any focused app.VocaHQ
VOZEB PRO is a source-available AI creation suite that combines a unified creative agent, canvas, short-drama production pipeline, asset library, and commercial admin backend in one Next.js application.csyqlz
A self-hosted AI app builder: describe an app, and a coding agent builds it inside an isolated sandbox on your own server, live at its own preview URL. You keep the code, data and infrastructure.tastyeffectco
An AI slide studio that runs inside Codex, turning prompts, repos or files into image-native decks with outline, style, editing, presenting and PPTX/PDF export.nexu-io
A CLI that applies plain-English instructions to your codebase via OpenAI models, writing changes directly to referenced files for refactoring and code generation.ferrislucas
ChatWeb crawls web pages and parses PDF, DOCX and TXT files, embeds the text with OpenAI's embedding API and stores it in a vector database so you can ask questions or get summaries based on the document content.SkywalkerDarren
TypeNo is a minimal macOS voice-input app. It transcribes speech locally and pastes the text into the active app in under a second, with no accounts or cloud dependency.marswaveai
A local-first multi-provider image and video generation studio for GPT, Grok, Gemini, NovelAI and ComfyUI, with browser, desktop, CLI and node-graph workflows.lidge-jun
A Windows chat assistant that reads conversations with local OCR, uses LLMs to judge intent and drafts reply candidates, while the user always sends manually.jev-chat
VoxtMini is a macOS voice-input app that runs local speech and language models: press a shortcut, speak, and transcribed, translated or rewritten text appears at your cursor.hehehai
Surf is a Next.js app that pairs OpenAI's computer-use model with an E2B virtual desktop, letting you give plain-language instructions and watch an AI carry them out on a cloud computer.e2b-dev
agent-qa is an AI QA engineer that runs natural-language end-to-end tests across web, Android, and iOS, keeping evidence and learning from each run through reviewable memory.vostride
Yogurt is a private AI meeting-notes desktop app. It records and transcribes calls locally with Whisper and summarizes them using an on-device model, with no account required.FrigadeHQ
WhisperSubTranslate is a local desktop app that generates SRT subtitles from video audio with whisper.cpp and translates them offline using Hy-MT2 or online engines.Blue-B
A desktop AI subtitle app that turns a video link or local file into finished captions. It automates downloading, transcription, translation, speaker labeling, timing alignment and burn-in, with a built-in editor and bilingual export.corvo007
Memex is an open-source, local-first AI journal for iOS and Android. It captures fragments of life and uses multi-agent AI to organize them into timeline cards, knowledge bases and insights.memex-lab
A browser extension that supercharges DeepSeek Chat with tool tags, file generation, in-browser code execution, persistent memory, MCP tools and custom skills, while keeping data local.EdgeTypE
A local-first desktop notification command center that aggregates work messages and stages approval-ready action cards with autonomous AI agents for review.aayushch
A pure-Rust implementation of Mistral's Voxtral realtime speech recognition and text-to-speech models, running natively with Burn or client-side in the browser via WASM and WebGPU.TrevorS
A Gradio desktop app that uses multiple AI source-separation models to split songs and videos into vocals, instrumentals and other stems, with batch processing, ensembles and easy stem downloads.Eddycrack864
A CrewAI multi-agent pipeline that turns a paper or topic into a cited commercialization screening report covering readiness, manufacturing, patents and market.shuxiachai
A research system for open-ended deep research that separates exploration from synthesis with critic, generator, writer and render agents, turning queries into grounded reports, posters and slide decks.agentdisco-project
Lapian Notes turns a film into an editable shot-by-shot study notebook, using AI to generate story-lane timelines, structure trees and audience-emotion curves while you watch.bkingfilm
An Obsidian plugin that turns notes and PDFs into a linked LLM-powered wiki, with graph-based retrieval, entity pages, and support for many model providers.GD4AI
MacParakeet is a fast, private, local-first voice app for Apple Silicon Macs offering system-wide dictation, file and media transcription, and meeting recording with on-device speech recognition.moona3k
A self-hosted meme search engine that uses vision models to describe images, builds vector and keyword indexes, and lets you find memes by content or text.meme-search
Meme Search is a self-hosted AI meme search engine that captions images with vision-language models and indexes them with vector embeddings for semantic and keyword search.neonwatty
ComfyUI nodes and workflows turning Krea 2 into an instruction-based, identity-preserving image editor, using dual source-image injection for consistent people and scene edits.lbouaraba
Nova3D is an AI design workspace that generates editable, part-aware 3D models and scenes from text or reference images. It supports parametric editing, physics-based simulation, and export to CAD, game engines, and robotics simulators.raresense
Recognize is a Nextcloud app that automatically tags photos, videos and music using on-premises face, object, landmark, action and music-genre recognition models.nextcloud
A Gradio-based voice cloning and design studio supporting multi-speaker dialogue, voice conversion, training, transcription, and sound effects.franckyb
A collection of more than fifty practical ComfyUI extension nodes for image workflows, supporting batch folder loading, DPI-controlled saving, mask analysis, portrait retouching, and photo layout.goohai
A cross-platform desktop voice-typing app that transcribes speech and polishes it with an LLM, adding rewriting, translation, and one-shot voice Q&A in any app.tover0314-w
Shotgun reads your whole repository, plans a feature, and writes codebase-aware specs split into staged, file-by-file instructions that Cursor, Claude Code, or Codex can follow, so large changes don't derail. It is a CLI/TUI with research, spec, plan, and export phases.shotgun-sh
OpenChatBI is an intelligent chat-based business intelligence tool. Powered by large language models, LangGraph, and LangChain, it lets users query, analyze, and visualize data through natural language and generated SQL.zhongyu09
A Rust TUI application for browser agents that pairs a custom LLM harness with real Chrome CDP control, letting you run, watch, and steer browser automation tasks from the terminal.browser-use
A desktop AI knowledge management app for macOS and Windows that captures clipboard content, lets AI organize it into a personal wiki and knowledge graph, and produces weekly insight reports.kdsz001
A local-first AI browser where people and multiple agents share real tabs, logged-in sessions, and page actions, with agents connecting over MCP or HTTP.hydro13
A stateless command-line LLM assistant: qq answers quick shell questions, while qa performs a single tool step such as reading a file or running a command, with support for many model providers.iagooar
A real-time person-intelligence app that snaps a face with smart glasses, fans out browser agents across social sources, and streams back a live dossier.affaan-m
ComeCut is a free, lightweight AI-powered video editing suite for web and desktop, with subtitle translation, dubbing, speech recognition and an AI editing assistant.juntaosun
ComfyUI-VibeVoice is a custom node pack that brings Microsoft's VibeVoice multi-speaker TTS and ASR models into ComfyUI workflows, handling model download and VRAM management.wildminder
A nutrition-advice AI system combining a LangGraph chat agent, RAG over health guidelines, YOLO food-image recognition, and research on version-aware retrieval, deployable with Docker Compose.archie0732
GitFriend is an AI-powered GitHub companion web app. It offers chat help with Git and GitHub questions, generates professional README files, and helps add Gitmoji to commit messages, using Firebase auth and a Groq-hosted model.krishn404
An AI-assisted reconnaissance and vulnerability-analysis toolkit that runs nmap, DNS, subdomain, JWT, and PCAP scans, then uses one or more LLMs to produce structured pentest reports via CLI or GUI.morpheuslord
An open-source text-to-speech document reader for EPUB, PDF, TXT, Markdown and DOCX. It offers synchronized read-along playback, word highlighting, audiobook export, and can run as a desktop app or self-hosted server.richardr1126
A desktop Python app that extracts transcripts from YouTube playlists or single videos and uses Google Gemini to turn them into structured, readable Markdown in a chosen style and language.Ebrizzzz
A Navidrome plugin that connects to an AudioMuse-AI server to add sonic-similarity instant mixes, artist radio, and similar-artist recommendations to the music frontend.NeptuneHub
An AI-powered karaoke authoring tool that separates vocals, transcribes lyrics and detects pitch to generate UltraStar files, MIDI and sheet music from any song.rakuri255
A CLI that takes a plain-English dev-tool idea, searches 19 package registries plus GitHub and Hacker News, ranks prior art by semantic similarity, and has an LLM judge how crowded the space is.r14dd
The Nietzschian Debugger is a command-line AI debugging companion that answers bugs with questions instead of fixes, challenging assumptions until you find the root cause yourself.DilawarShafiq
A Streamlit app that loads arXiv papers into a Redis vector index and answers grounded questions, built as a learning project for academic RAG workflows.redis-developer
A research system that turns a single image into a fully rigged, animation-ready 3D asset, shipping inference code, a Gradio demo and pretrained models for 3D generation, 4D reconstruction and video editing.vast-ai-research
An LV2 plugin that runs neural amp models through the NeuralAudio engine, bringing machine-learning guitar amp and pedal tones into LV2 hosts.mikeoliphant
A full-stack local RAG demo for chatting with PDFs. It combines Ollama, LangChain and ChromaDB with Next.js, Streamlit and FastAPI interfaces, returning answers with citations from uploaded documents.tonykipkemboi
A RAG system for querying and summarizing large, heterogeneous document collections. It supports PDFs, web pages, audio, video and Anki decks with many LLM providers, returning sourced answers and structured summaries.thiswillbeyourgithub
A desktop invoice assistant that collects e-invoices from email, extracts data with OCR and vision models, files them automatically, and produces an Excel reimbursement summary.EthanYoQ
A macOS floating assistant that reads the active chat, uses a local model to judge intent and risk, and suggests reply candidates in several tones without ever sending.jev-chat
Turns the documentation a team already writes into runnable end-to-end tests, flagging where the product and docs disagree, with a web dashboard and pull-request checks.truecourse-ai
A local-first macOS dictation app: hold a shortcut, speak, and on-device Whisper, Parakeet, or Apple speech transcribes your words into any app.opennookorg