Fresh & distinctive

Deep scholarly Quran study tool: word-by-word morphology, i'rab grammar, dependency treebanks, and an AI assistant that answers in plain English fully offline.

Offline LLM client that runs GGUF/MLX/CoreML/Apple Foundation models from Hugging Face, with on-device RAG over your PDFs/EPUBs, vision, and optional remote endpoints.

Text-native audiobook studio that clones a voice from a 3-10s clip, auto-assigns speakers, and generates expressive narration entirely on-device, no subscriptions.

Converts 2D clips into spatial (3D) video for Apple Vision Pro using custom on-device ML models; non-destructive timeline editor, fully private.

On-device TTS and voice cloning powered by Qwen3-TTS via Apple MLX; clone a voice from a short clip across 11 languages, fully private.

Extracts hardcoded/burned-in subtitles from video into an editable .srt using on-device AI detection and multi-language OCR; nothing leaves the device.

Voice-first notes that Whisper-transcribe and auto-compile into a living personal wiki of topics, people, and projects; persona-tunable, private by design.

Turns live news into structured intelligence dashboards (timelines, forecasts, actor networks) using Apple Intelligence entirely on device.

Private speech coach that runs STT, language reasoning, and voice synthesis locally on the Neural Engine, scoring filler words, pace, and clarity in real time.

Private knowledge base that ingests many file types and lets you chat and semantically search them with a local LLM (any GGUF model); 100% offline.

Self-generating sequencer/synth whose 'genome' you sculpt by typing a prompt like 'warm evolving pad' via on-device AI; AUv3 plugin, nothing leaves device.

Local music player and full mobile DAW with a Neural Amp Modeler guitar amp, AUv3 hosting, synth, and drum machine, plus on-device AI mix chat; no subscription.

Fast, well-loved Mac duplicate cleaner for huge photo libraries that detects near-duplicates by content and adds AI-powered quality scoring to auto-keep the best.

Native Mac DAW with 100% on-device AI: CoreML Demucs/Open-Unmix stem separation, voice isolation, Whisper-style transcription, and 54-voice Kokoro TTS.

Generative music composer built on the RWKV LM that creates new pieces or continues your melodies entirely locally, with auto-chords and no internet needed.

Local TTS studio with 5-second voice cloning, 318+ voices and a multitrack editor, powered by the Chatterbox neural model — runs fully offline on Mac.

Pro Mac cutout tool with four on-device matting models (Apple Vision, IS-Net DIS, MODNet, BiRefNet), edge control and batch export.

Pulls subtitles from any video to .srt, including image-based and burned-in tracks read frame-by-frame via on-device Apple Vision + Tesseract OCR.

Turns single photos and videos into red-cyan anaglyphs, side-by-side, and Vision Pro spatial 3D using on-device ML depth prediction.

Turns your Mac into a private local AI server: run 500+ language/vision models, chat in a built-in web UI, and expose them over your network via MCP and local APIs.

Apple's pro DAW with on-device Stem Splitter that separates a mix into six parts, plus AI Session Players and a Mastering Assistant.

Apple's flagship pro image editor with machine-learning Super Resolution, Denoise, ML Enhance, background removal and the Repair object-removal tool, all Metal-accelerated.

Offline Mac transcriber using Whisper large-v3-turbo plus a local Qwen model to summarize, translate and answer questions about your transcripts; pay-once.

Local LLM workstation for Mac: MLX inference with GPU acceleration, an OpenAI-compatible server, multi-Mac distributed inference, and on-device RAG.