owsm-v4-medium-1b (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
owsm-v4-medium-1b.gguf	3899.3 MB	cfaf152f45ae7e6508d3d054232c3e81c6b5f77ddf697e964258a3c0b440ec0b

Usage

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o owsm-v4-medium-1b.gguf \
  https://huggingface.co/vokra/owsm-v4-medium-1b/resolve/main/owsm-v4-medium-1b.gguf

vokra-cli run --model owsm-v4-medium-1b.gguf --input input.wav

Provenance

Field	Value
Architecture	owsm-v4-medium-1b
Tensors	1172
Upstream source	espnet/owsm_v4_medium_1B (Open Whisper-style Model, ESPnet fully-open Whisper alternative, cc-by-4.0)
Upstream licence	cc-by-4.0
Licence class	attribution-required
Registry model id	owsm-v4-medium-1b
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 cc-by-4.0, unchanged from upstream. Conversion does not alter the licence, and your obligations run to the upstream author.

Attribution required

This application uses the ESPnet OWSM v4 medium 1B model (Open Whisper-style Model, fully-open Whisper alternative trained on 320k h of multilingual speech). Model weights are licensed under CC-BY 4.0 (attribution required; commercial use permitted). Copyright (c) ESPnet Community. Source: https://huggingface.co/espnet/owsm_v4_medium_1B

This licence obliges you to display the attribution above when you ship something built on these weights. Vokra surfaces it at runtime via vokra_model_attribution (C ABI) and a CLI banner.

Verifying this file

shasum -a 256 owsm-v4-medium-1b.gguf
# expect: cfaf152f45ae7e6508d3d054232c3e81c6b5f77ddf697e964258a3c0b440ec0b