§ Privacy posture

Privacy + security

What leaves your Mac and what does not. Speech-to-text runs locally by default, hosted providers are opt-in with bring-your-own-key credentials in Keychain, and retention stays under your control.

§ Pages4 pages

Cloud providers
Cloud providers (BYOK)
Hosted STT, polish, and meeting analysis are available behind a Bring-Your-Own-Key model. OpenAI, Anthropic, or any OpenAI-compatible endpoint (Ollama, vLLM, LM Studio, a self-hosted gateway). Keys live in Keychain. Off by default.
Opt-in
Local AI
Local AI models
Speech-to-text runs through Whisper (via the WhisperKit engine) or Parakeet (via the FluidAudio engine), pick per task. Enhanced polish runs through our own bespoke MLX model, fine-tuned for cleaning spoken input rather than general chat; meeting analysis runs through bundled MLX too. Advanced users can point Dettivo at compatible local MLX folders or cached Hugging Face MLX repos.
Local-only
Privacy
Privacy posture
Dictation and meeting capture run locally on Apple Silicon. Hosted providers (OpenAI, Anthropic, OpenAI-compatible) are opt-in, BYOK, and credentialed in Keychain. No telemetry. No bot joins meetings. You control retention.
Local-only
STT engines
STT engines
Dictation and meetings can pick different STT models. Whisper tiers (via the WhisperKit engine) and Parakeet (via the FluidAudio engine) ship by default, both run on-device on Apple Silicon. Cloud STT is opt-in, BYOK, and currently used for dictation paths only.

§ Other categories

Cloud providers (BYOK)

Local AI models

Privacy posture

STT engines