Matt Mireles

Fine-tune Gemma 4 and 3n with audio, images and text on Apple Silicon, using PyTorch and Metal Performance Shaders.

PyTorch → CoreML conversion pipeline for Kokoro TTS. Unlocks fast on-device text-to-speech on Apple Neural Engine.

transform an entire GitHub repo into a simple text file so that you can upload it to an LLM (Claude, ChatGPT, etc.)

Google's MRT2 → Core ML, full Apple Neural Engine execution, zero GPU. Runs locally on iPhone without burning your hand.

All Your AIs. All At Once. The ultimate  macOS desktop app for multi-GPT chat.

Run a local coding model in Cursor chat & agent on macOS. Load it when you need it; free memory when you don't.

Record audio in the browser. Get an MP3 you can feed to an LLM.

Hit Gemini API's secret 20GB file limit? Dump them all and get back to work.

Interactive reading tool that highlights words as you listen

A text-to-speech (TTS), speech-to-text (STT) and speech-to-speech (STS) library built on Apple's MLX framework, providing efficient speech analysis on Apple Silicon.

An editor that converts images to pixel art, add dither, set palettes, and more! Made entirely in p5js.

Signal Static — Outpost 7: Ursina FPS with story, practice range, and LAN co-op

Swift port of Misaki G2P (grapheme-to-phoneme) library that can be used e.g. to generate phonemization for Kokoro text-to-speech engine

Magenta RealTime 2: An Open-Weights Live Music Model

No description.

Label Studio is a multi-type data labeling and annotation tool with standardized output format

✨✨VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction

Zerg Autonomous AI Manifest Suite

OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training

Experiment codebase for DiLoCo-SWARM

EXO Gym is an open-source Python toolkit that facilitates distributed AI research.

NeMo text processing for ASR and TTS - re-implemented in Swift

ConMamba for Automatic Speech Recognition

Kokoro TTS for iOS and macOSX using MLX Swift

a better future for multimodal content

No description.

No description.