ggml-org/llama.cpp

GGUF convert support for Vibevoice

Open

#17,488 建立於 2025年11月25日

在 GitHub 查看
 (3 留言) (1 反應) (0 負責人)C++ (110,169 star) (18,202 fork)batch import
3rd partyenhancementgood first issuehelp wantedmodeltts

描述

Hey there! I am trying to convert my fine-tuned vibevoice into a gguf format for inference but cant find proper code or converter any where. The architecture is currently unsupported to any known converter like llama.cpp but I have already seen many variation of gguf-formated vibevoice. Can anyone help me with that?

貢獻者指南

技術棧
cpp
領域
backendmachine learning
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
5
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
over 1 week
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
needs maintainer response
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
unclear
前置要求
無
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
10
研究方向
This issue requests support for the 'Vibevoice' model architecture in GGUF format. To implement, one must first understand the Vibevoice model architecture and its compatibility with the GGUF format. Existing conversion scripts in the llama.cpp repository (like convert.py) should be studied as reference. The model's architecture may require adding a new model type in the C++ inference code. There are no linked PRs or comments providing further details, so the contributor should start by researching the architecture and proposing a design.