Audio8-TTS-Preview-0.6b needs about 2.1 GB of memory at Q8_0 (a 1.4 GB file) with an 8K context window, so it fits on a 4 GB GPU.
Audio8 TTS Preview 0.6B – compact multilingual text‑to‑speech with zero‑shot voice cloning
Audio8 TTS Preview 0.6B is a 1.3 B‑parameter DualAR model for multilingual speech generation. It supports zero‑shot voice cloning and runs on a CUDA‑capable GPU. The model is released under Apache‑2.0 and is available in GGUF format.
Good at
- Zero‑shot voice cloning
- Multilingual TTS in 11 languages
- Compact 1.3 B parameter size
Limits
- No vision support
- Requires CUDA GPU for best performance
- Limited language coverage in preview
Best for
- Creating synthetic voice recordings
- Voice cloning for short clips
- Rapid prototyping of multilingual TTS
Who it’s for. Developers and researchers needing a small, multilingual TTS model with voice cloning. The model is suitable for experimentation and low‑resource deployment on GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 1.4 GB | 2.1 GB | Fits | Get file → |
The Q8_0 quantization is the default and requires about 1.43 GB of storage, with an estimated 2.1 GB VRAM at 8 k context. For lower VRAM budgets, consider a lower‑precision quant if available, but only Q8_0 is listed.
What GPU do you need for Audio8-TTS-Preview-0.6b?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q8_0 (1.4 GB file, about 2.1 GB in use).
Which Audio8-TTS-Preview-0.6b quant should you choose?
Q8_0 (1.4 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Audio8-TTS-Preview-0.6b
How much VRAM does Audio8-TTS-Preview-0.6b need?
About 2.1 GB at Q8_0 with an 8K context window. That is the 1.4 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Audio8-TTS-Preview-0.6b on an 8 GB GPU?
Yes, with Q8_0: the 1.4 GB file needs about 2.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Audio8-TTS-Preview-0.6b GGUF?
It depends on the quant: Q8_0 is 1.4 GB.
Where can I download Audio8-TTS-Preview-0.6b GGUF?
The files are on Hugging Face in the js-byte/Audio8-TTS-Preview-0.6b-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Audio8-TTS-Preview-0.6b use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Audio8 TTS Preview on 8 GB VRAM?
Yes, the Q8_0 quant uses about 2.1 GB VRAM at 8 k context, well within an 8 GB GPU.
Does this model support image input?
No, Audio8 TTS Preview is a text‑to‑speech model and does not process visual data.
Is the model multilingual?
Yes, it supports 11 languages: Cantonese, Chinese, Dutch, English, French, German, Italian, Japanese, Korean, Polish, Spanish.