- Who it helps
- Local audio experimenters who already manage model files.
- Availability
- Release page dated October 31 at 04:51 lists downloadable templates. The no-language-model YuE2 template was retrieved as text; installation and inference were not run.
Music choices
The release offers YuE2 with supplied lyrics, a small lyrics model or a large lyrics model. These configurations have different resource needs; the largest is described for large GPUs.
Speech and transcription
Ported speech models have different voice-cloning and instruction support. MusicUI also adds a Whisper transcription tab, useful when a speech model requires reference text.
What you can do
- Read the primary evidence and its stated access limits before designing a trial. Open reference ↗
Where the evidence stops
Model licenses, downloads and hardware suitability need checking for each selected model. No audio-quality or speed comparison was performed.
Sources read. Sources checked October 31, 2026 UTC. No third-party software, model inference or complete workflow was executed.
Sources & further reading
GitHub · official
KoboldCpp 1.124 adds more local music and speech models ↗
Primary documentation checked; no workflow executed or reported result reproduced.
LostRuins / KoboldCpp · page checked 2026-10-31 · Published 2026-10-31
Coverage note: Independent attention and end-to-end execution remain unverified.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
KoboldCpp 1.124 integrates audio.cpp model support into its music and speech interface.