Whispall mirror — Whisper large-v3-turbo GGML

Availability mirror of the exact ASR model files used by Whispall (local, privacy-first macOS dictation & assistant app) via whisper.cpp.

File	sha256	Role in Whispall
`ggml-large-v3-turbo-q8_0.bin`	`317eb69c11673c9de1e1f0d459b253999804ec71ac4c23c17ecf5fbe24e259a1`	Default dictation + meetings model
`ggml-large-v3-turbo.bin`	`1fc70f774d38eb169993ac391eea357ef47c88757ef72ee5943879b7e8e2bc69`	Advanced install path (full precision)

Mirrored from: ggerganov/whisper.cpp (upstream revision 5359861c739e955e79d9a303bcbc70fb988958b1 at mirror time); both files byte-identical to the upstream LFS objects.
Purpose: hosting stability for Whispall installs — the upstream repo tracks a moving main; this mirror freezes the exact files the app expects.

License

OpenAI Whisper weights and the whisper.cpp GGML conversions are MIT-licensed. © OpenAI (model), conversions by Georgi Gerganov / the whisper.cpp project.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support