Modelpublic

vokra/whisper-medusa-v1

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes171downloads

Model Card

whisper-medusa-v1 (Vokra GGUF)

Converted to the Vokra GGUF format for Vokra, a zero-dependency speech-AI inference runtime.

This is a conversion, not a new model. The weights are the upstream ones; Vokra re-packages them so its runtime can memory-map them directly. Credit for the model belongs upstream — see Source below.

Files

File	Size	SHA-256
`whisper-medusa-v1.gguf`	5956.2 MB	`af67b0beb21b0634b8831d31158e33e5b9703ba3215573a196bf65558d949b28`

Usage

bash

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o whisper-medusa-v1.gguf \
  https://huggingface.co/vokra/whisper-medusa-v1/resolve/main/whisper-medusa-v1.gguf

bash

vokra-cli run --model whisper-medusa-v1.gguf --input input.wav

Provenance

Field	Value
Architecture	`whisper-medusa-v1`
Tensors	1281
Upstream source	aiola/whisper-medusa-v1 (Whisper backbone + Medusa speculative-decoding heads, apache-2.0 per ticket header; primary-source sign-off pending in docs/license-audit.md §3.1)
Upstream licence	`apache-2.0`
Licence class	`permissive`
Registry model id	`whisper-medusa-v1`
Vokra GGUF schema	1
Converted by	vokra-core 0.1.0-alpha.0

Every row above is read out of this file's own vokra.* metadata, so the card cannot claim something the artifact does not carry.

Licence

The weights are distributed under apache-2.0, unchanged from upstream. Conversion does not alter the licence, and your obligations run to the upstream author.

Verifying this file

bash

shasum -a 256 whisper-medusa-v1.gguf
# expect: af67b0beb21b0634b8831d31158e33e5b9703ba3215573a196bf65558d949b28