Explore our curated selection of highly-optimized offline AI transcription engines, rigorously tested and built to run natively on Apple Silicon (M1/M2/M3/M4) devices.
Voxtral Mac Large
The flagship offline real-time model.
VoxBar Kyutai 2.6B (6GB VRAM)
Meeting and recording transcription (~4s delay).
Nemotron Speech ASR 0.6B
NVIDIA Nemotron Speech model.
Faster-Whisper Large
Maximum accuracy Whisper variant.
Distil-Whisper
Distilled Whisper for faster execution.
Faster-Whisper (CTranslate2)
The classic Whisper engine optimized with CTranslate2.