granite-speech-4.1-2b (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
model.gguf	4608.4 MB	7b073cb412b0c6474a2fa6b44562e90f26f64cdb9a7e960a21292bcebe6b7fef

Usage

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o model.gguf \
  https://huggingface.co/vokra/granite-speech-4.1-2b/resolve/main/model.gguf

vokra-cli run --model model.gguf --input input.wav

Provenance

Field	Value
Architecture	granite_speech
Tensors	940
Upstream source	ibm-granite/granite-speech-4.1-2b (Conformer CTC encoder + Granite-4.0-1b-base LLM decoder + BLIP-2 q-former projector, apache-2.0)
Upstream licence	apache-2.0
Licence class	permissive
Registry model id	granite-speech-4.1-2b
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

shasum -a 256 model.gguf
# expect: 7b073cb412b0c6474a2fa6b44562e90f26f64cdb9a7e960a21292bcebe6b7fef