Meetily
zackriya-solutions/meetily
100% local AI meeting assistant built with Tauri, Rust, Whisper, and Ollama for private transcription and summaries.
SEARCH RESULTS
zackriya-solutions/meetily
100% local AI meeting assistant built with Tauri, Rust, Whisper, and Ollama for private transcription and summaries.
genspark-ai/genoffice
Open-source offline AI office suite featuring byte-preserving round trips for Word, Excel, and PowerPoint.
ace-step/ace-step-1.5
Free, open-source (MIT) AI music model that runs on your own PC: describe a song or paste lyrics and get a full track with instruments and vocals. A portable package for Windows, a browser interface and no account. An NVIDIA card is recommended; Vietnamese lyrics are untested.
unslothai/unsloth
Ultra-fast, memory-efficient LLM fine-tuning library saving 70-80% VRAM with manual Triton autograd kernels.
jamiepine/voicebox
Free, open-source (MIT) AI voice studio that runs on your own computer: clone a voice from a short sample, turn text into speech with several engines and type by voice. Installers for Windows and macOS, no account needed. Vietnamese is not among the speech languages the README names.
debpalash/voicestudio
Open-source, 100% local ElevenLabs alternative for voice cloning, dubbing, and dictation supporting 646 languages.
pnnbao97/vieneu-tts
Open-source (Apache-2.0) Vietnamese text-to-speech built by a Vietnamese developer that runs on your own machine. The open v3 Turbo model outputs 48 kHz audio with 25 preset voices covering Northern, Central and Southern accents, handles mixed Vietnamese-English text, supports emotion cues such as [cười] (laugh) and [thở dài] (sigh), and clones your own voice from a 3-8 second clip. It runs about 2x real-time on a plain CPU with no GPU needed, and ships a local web UI, a Python SDK and an OpenAI-compatible speech API.
justvugg/colibri
Pure C inference engine designed to run massive 744B to 2.8T MoE models on consumer hardware.
lyogavin/airllm
Run 70B large language models on consumer 4GB GPUs using layer-by-layer disk streaming inference.
fujiwarachoki/moneyprinterv2
Automated short-form video creation engine integrating local Ollama scripts, TTS audio, subtitles, and stock footage.
significant-gravitas/autogpt
A team of AI experts you brief in plain English: Otto coordinates specialists for content, prospect research and admin work, with a drag-and-drop workflow builder and a marketplace. Self-host it free with Docker using your own model keys or a local model through Ollama.
juliusbrussee/caveman
A skill plus a local proxy that makes AI coding agents say less and read less: 33.2% fewer input tokens (pinned 54-run Claude Code suite) and 8.5% fewer output tokens per JetBrains' independent A/B, with no detectable quality change. Works with 30+ agents, one command, no account or API key.
thu-maic/openmaic
Open-source (MIT) platform from Tsinghua University: type a topic or upload a document and AI builds a whole interactive classroom with lesson pages, quizzes, simulations and AI teachers and classmates. Exports pptx or html. Self-host with Docker; needs an AI service API key (usually pay-per-use) or a local model.
pascalorg/editor
Open-source, local-first 3D architectural building editor built with React Three Fiber and WebGPU. Features a local CLI and authenticated MCP service connecting AI agents (Claude Code, OpenAI, Codex) to inspect 3D scenes, perform structural exploded views, section cuts, MEP X-ray inspection, and bounded furniture fitting.
localsend/localsend
Free and open-source alternative to AirDrop: share files and messages between nearby devices over your local network, with no internet connection and no third-party servers. Runs on Windows, macOS, Linux, Android and iOS; files are transferred as-is (no chat-app recompression), folders are sent recursively, traffic is HTTPS-encrypted and no account is needed. Install via winget, Homebrew, Flatpak, the App Store or Play Store.
fastapi/full-stack-fastapi-template
The FastAPI team's own full-stack project template: FastAPI + SQLModel + PostgreSQL on the backend, React + TypeScript + Vite + Tailwind CSS + shadcn/ui on the frontend. It ships JWT login, secure password hashing, email password recovery with Mailpit for local testing, Pytest and Playwright tests, GitHub Actions CI/CD, and Docker Compose with Traefik for automatic HTTPS when self-hosting. Hit "Use this template" on GitHub and you start from a repo that already runs.
niko1221/strata
Free, open-source (MIT) program that runs a very large AI model (Qwen3.8-Flash-Next, about 125B parameters) offline on a gaming PC with an NVIDIA or AMD card of 12 GB+ VRAM and 32 GB+ RAM. Browser chat, installed with one .bat file on Windows. Heavy: about 80 GB of free disk.
jasonzhu1207/zenconverter
Free, open-source (AGPL-3.0) Android app for converting files on your phone: merge, compress and password-protect PDFs, images to PDF, image formats, video compression and merging, audio extraction, subtitles and fonts. No ads, no account, processed locally. Needs Android 8.0 or newer, no Vietnamese interface yet.
storytold/photocraft
Free, open-source (MIT or Apache-2.0) Photoshop alternative for Windows, macOS and Linux with layers, masks, adjustment layers, text, brushes and real PSD support. Runs locally with no account. Early alpha, no Vietnamese interface yet.
storytold/pdfcraft
Free, open-source (MIT or Apache-2.0) PDF reader and editor in the style of Adobe Acrobat for Windows, macOS and Linux. Search, organize pages, merge and split PDFs, comments, forms, passwords and printing. Runs locally with no account or telemetry. Alpha software, no Vietnamese interface yet.
storytold/filmcraft
Free, open-source (MIT or Apache-2.0) Premiere Pro alternative for Windows, macOS and Linux with a multi-track timeline, transitions, colour grading, audio mixing and H.264 MP4 export. Runs locally with no account. Young software, about 50-60% ready for real work by the project's own estimate.
storytold/lightcraft
Free, open-source (MIT or Apache-2.0) Lightroom-style photo manager and RAW editor for Windows, macOS and Linux. Photo library, ratings, RAW development, masks, presets and batch export. Runs locally with no account or subscription. Young project, no Vietnamese interface yet.
jub0t/concat
Free, open-source (AGPL-3.0) video editor and CapCut alternative for Windows, macOS, Linux and Android. According to the project: no watermark, no account, and your footage is never uploaded. Auto-captions, text-to-speech, background removal, 170+ effects and 4K export. Still in beta; no Vietnamese interface yet.
upscayl/upscayl
Free, open-source (AGPL-3.0) AI image upscaler that runs on your own computer. Pick a small, blurry or pixelated picture, press one button and get a bigger, sharper copy. Seven built-in upscaling styles and batch folders. Windows, macOS and Linux, no account needed; requires a Vulkan-compatible graphics card.
nightly-labs/openbot
A local-first desktop workspace for persistent AI teammates. Run Codex, Claude, and Grok with dedicated workspaces, task queues, file sharing, browser control, and agent-to-agent collaboration.
comfy-org/comfyui
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
graphify-labs/graphify
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
home-assistant/core
🏡 Open source home automation that puts local control and privacy first.
floci-io/floci
Light, fluffy, and always free - The AWS Local Emulator alternative
anomalyco/hex
Local voice dictation for mac, windows and linux