Serveurperso/chatterbox-GGUF overview
Chatterbox GGUF GGUF conversions of Chatterbox turbo and multilingual v3 for the llama.cpp mtmd TTS pipeline. Proof of concept built with @ngxson, under review…
Runs locally from ~295.7 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| chatterbox-GGUF-BF16.gguf | GGUF | BF16 | 1002.1 MB | Download |
| chatterbox-GGUF-Q4_K_M.gguf | GGUF | Q4_K_M | 303.0 MB | Download |
| chatterbox-GGUF-Q8_0.gguf | GGUF | Q8_0 | 532.6 MB | Download |
| chatterbox-turbo-GGUF-BF16.gguf | GGUF | BF16 | 833.3 MB | Download |
| chatterbox-turbo-GGUF-Q4_K_M.gguf | GGUF | Q4_K_M | 295.7 MB | Download |
| chatterbox-turbo-GGUF-Q8_0.gguf | GGUF | Q8_0 | 459.3 MB | Download |
| mmproj-chatterbox-GGUF-F16.gguf | GGUF | F16 | 567.7 MB | Download |
| mmproj-chatterbox-GGUF-Q8_0.gguf | GGUF | Q8_0 | 371.2 MB | Download |
| mmproj-chatterbox-turbo-GGUF-F16.gguf | GGUF | F16 | 528.3 MB | Download |
| mmproj-chatterbox-turbo-GGUF-Q8_0.gguf | GGUF | Q8_0 | 329.9 MB | Download |
Model Details
| Model ID | Serveurperso/chatterbox-GGUF |
|---|---|
| Author | Serveurperso |
| Pipeline | — |
| License | mit |
| Base model | ResembleAI/chatterbox,ResembleAI/chatterbox-turbo |
| Last modified | 2026-08-03T07:51:00.000Z |
Model README
---
license: mit
base_model:
- ResembleAI/chatterbox
- ResembleAI/chatterbox-turbo
---
Chatterbox GGUF
GGUF conversions of Chatterbox turbo and multilingual v3 for the llama.cpp
mtmd TTS pipeline. Proof of concept built with @ngxson, under review:
https://github.com/ngxson/llama.cpp/pull/107
Requires a build of the PR branch. Voice cloning takes any reference clip
via --tts-speaker-file, sampling defaults per variant are embedded in the
GGUFs (general.sampling.*).
llama-tts -m chatterbox-turbo-GGUF-Q8_0.gguf \
-mm mmproj-chatterbox-turbo-GGUF-Q8_0.gguf \
--tts-speaker-file ref.wav -p "Hello there." -o out.wav
Multilingual prompts carry a language tag: -p "[fr]Bonjour."
Run Serveurperso/chatterbox-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models