Meetily: Privacy-first AI meeting assistant

Privacy-first AI meeting assistant that runs locally. Captures meetings, transcribes in real-time, and generates summaries without cloud. Supports macOS and Windows, with flexible AI providers like Ollama, Claude, Groq, and

Overview

Meetily is a privacy-first AI meeting assistant that runs entirely on your local machine. It captures meetings, transcribes them in real-time, and generates summaries without sending data to the cloud. Built on Rust with Tauri, it supports macOS and Windows, with flexible AI provider options including Ollama, Claude, Groq, and OpenAI-compatible endpoints.

Key Features

Local First: All processing is done on your machine. No data ever leaves your computer.
Real-time Transcription: Get a live transcript of your meeting as it happens.
AI-Powered Summaries: Generate summaries of your meetings using powerful language models.
Multi-Platform: Works on macOS, Windows, and Linux.
Open Source: Meetily is open source and free to use.
Flexible AI Provider Support: Choose from Ollama (local), Claude, Groq, OpenRouter, or use your own OpenAI-compatible endpoint.

Use Cases

Professionals who need to maintain complete control over sensitive information.
Enterprises that need advanced meeting intelligence without compromising on privacy, compliance, or control.
Legal and healthcare providers who require data sovereignty and compliance with regulations like GDPR.
Teams and organizations (2-100 users) requiring self-hosted deployment.

Getting Started

Windows: Download the latest x64-setup.exe from Releases and run the installer.
macOS: Download meetily_0.4.1_aarch64.dmg, open it, drag Meetily to Applications, and launch.
Linux: Build from source following the provided guides (docs/building_in_linux.md and docs/BUILDING.md).

Deployment & Requirements

Windows: Requires an AVX2-capable x64 CPU; CUDA acceleration requires a source build with NVIDIA CUDA toolchain.
macOS: Requires Apple Silicon (aarch64) for the provided package.
Linux: Build from source with appropriate acceleration configuration.
GPU acceleration: macOS uses Metal and CoreML automatically; Windows packages use Vulkan; Linux requires source build.

Before You Adopt

License: MIT. Review its terms before using, modifying, or distributing the project.
Speaker diarization is not yet available in the Community Edition; it is planned for the PRO version.
CUDA acceleration is not enabled by default in the Windows installer; it requires a source build with a compatible NVIDIA CUDA toolchain.
The Windows installer requires an AVX2-capable CPU.

Meetily