MLX is a NumPy-like array framework designed for efficient and flexible machine learning on Apple silicon, brought to you by Apple machine learning research.
On-device, real-time multimodal AI. Have natural voice and vision conversations with an AI that runs entirely on your machine. Powered by Gemma 4 E2B and Kokoro.
MLX Local Serving (MLS) - Unified ASR, TTS, and Translation on Apple Silicon
Local AI mesh — multiple agents, shared persistent memory, real-time dashboard, browser automation. Runs on your hardware.
🤖 JARVIS — Privacy-first local AI voice assistant for macOS (Apple Silicon). Korean-first, English-supported. Powered by whisper.cpp, Qwen3, MLX, and LanceDB. Fully offline-capable.
A lightweight RAG (Retrieval-Augmented Generation) server built with Python, FastAPI and MLX for efficient local inference on Apple Silicon M chips.
🎙️ 100% Local AI Transcription with Speaker Diarization — No API key, no cloud, no cost. Supports 99+ languages, dual engine (CPU + Apple Silicon GPU), exports to SRT/TXT/DOCX.
MISSION_CONTROL // Live telemetry dashboard — 39 modules, 3 pillars, Apple Silicon native
Push-to-talk voice transcription for macOS using MLX Whisper. Beautiful UI, fast on-device AI, privacy first.
LLM eval framework. Compare any model via OpenAI-compatible API.
Dictate privately on macOS with open-wispr. Speak freely while keeping all audio and text on your device, ensuring your privacy and convenience.
Educational proof-of-concept: how model distillation works, demonstrated with Claude as teacher and Llama 3.2 on Apple Silicon (MLX). Companion to "Why the Best AI You'll Ever Have Unrestricted Access To Is the AI You Have Right Now"