# KoboldCpp v1.124 adds YuE2 music generation and dozens of TTS models

Digest AI · Agents & Tools · published 2026-10-31T05:24:06Z

Canonical: https://digestai.news/story/koboldcpp-v1-124-adds-yue2-music-generation-and-dozens-of-tts-models

## Summary

KoboldCpp 1.124 introduces a small subset of popular models from audio.cpp, expanding its built‑in Music Gen and TTS Gen capabilities. The release supports the YuE2 music generation model in three configurations – a lightweight version without an LLM, a small‑LLM option that adds a 4B language model for lyric creation (recommended), and a large‑LLM variant for high‑end GPUs.

The update also brings new image generation support via sd.cpp sync, adding models such as MOSS Image, Fibo 2 Edit, Flux 3 Klein, and updates to existing ones like Krea 2 Turbo and Qwen Image 2.1. AI Horde now supports Chat Completions workers, enabling jinja, reasoning, tool calls, streaming, and vision. Various performance and stability fixes round out the release, including CUDA‑graph enablement, parallel context handling, and improved SWA for Gemma4 models.

Users can launch the updated binary for Windows, Linux, macOS, or AMD‑Vulkan builds and access the MusicUI at http://localhost:5001/musicui, which now includes a transcribe tab for Whisper models and a directory of supported TTS models.

## Key points

- KoboldCpp 1.124 adds YuE2 music generation with optional 4B LLM for lyrics
- AI Horde now supports Chat Completions workers for jinja, tool calls, streaming and vision

## Why it matters

The expanded audio and vision models let developers and creators generate music, voiceovers, and images directly within KoboldCpp, lowering the barrier to multimodal content production.

## Sources

1. [Koboldcpp v1.124 released](https://github.com/LostRuins/koboldcpp/releases/tag/v1.124) (github.com, 2026-10-31, primary source)

## Cite

Digest AI, "KoboldCpp v1.124 adds YuE2 music generation and dozens of TTS models", 31 October 2026, https://digestai.news/story/koboldcpp-v1-124-adds-yue2-music-generation-and-dozens-of-tts-models

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/koboldcpp-v1-124-adds-yue2-music-generation-and-dozens-of-tts-models.json