Fact · Dictation software
11 = yes, 0 = no
The HoldSpeak README says the language model can be any OpenAI-compatible endpoint the user chooses.
Quote found word for word on the live source, last checked October 25, 2026
In the source's own words
“The language model is yours: a local GGUF or MLX model, or any OpenAI-compatible endpoint you choose.”
HoldSpeak README (GitHub repository), karolswdev, GitHub (karolswdev/HoldSpeak), 2026, Headline section.
Open the original- What it measures
- The paragraph under the banner, describing the model choice.
- What it does not show
- The page names no provider. The row exists only when the user assigns such an endpoint.
In the data
A row of Dictation App Features Compared: Offline, Privacy, AI, as HoldSpeak: third-party endpoint: 1 1 = yes, 0 = no.
- App
- HoldSpeak
- Feature
- Sends text to a third-party AI
- Detail
- OpenAI-compatible endpoint, if you choose one
- As of
- 2026-10
- Note
- 1 means the maker's page says the app does this