yue-xcodec-mini (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
yue-xcodec-mini.gguf	1726.2 MB	60e21aa5335646080102196454d7ffad5e012467d6f5eb9b776bf07d666b02bc

Usage

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

vokra-cli run --model yue-xcodec-mini.gguf --input input.wav

Provenance

Field	Value
Architecture	yue_xcodec_mini
Tensors	2145
Upstream source	m-a-p/xcodec_mini_infer (YuE xcodec-mini bundle = SoundStream RVQ codec 25 Hz + HuBERT-base semantic encoder + Vocos decoder head, apache-2.0)
Upstream licence	apache-2.0
Licence class	permissive
Registry model id	yue-xcodec-mini
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 yue-xcodec-mini.gguf
# expect: 60e21aa5335646080102196454d7ffad5e012467d6f5eb9b776bf07d666b02bc