MOSS-TTS-v1.5 needs about 15.3 GB of memory at Q8_0 (a 13.2 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 3 quants, from 9.3 GB (Q4_K) to 20.5 GB (BF16).
MOSS‑TTS‑v1.5 – a high‑fidelity audio tokenizer for TTS
MOSS‑TTS‑v1.5 is a GGUF package for audio.cpp that includes the 8B delay‑pattern backbone, the MOSS‑Audio‑Tokenizer v1 codec, tokenizer and model spec. It is fully self‑contained and can be cloned on a 24 GB RTX 3090. The model is licensed under Apache‑2.0.
Good at
- High‑fidelity speech synthesis
- Semantic‑rich audio tokenization
- Low‑latency streaming inference
Limits
- No vision support
- Requires 8 k context for memory estimates
- Large VRAM needed for bf16
Best for
- Voice cloning projects
- Audio generation pipelines
- Speech‑to‑speech conversion
Who it’s for. Developers building TTS or audio generation systems that need a compact, transformer‑based tokenizer. The model is suitable for users with a recent audio.cpp build and a GPU with at least 12 GB VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 9.3 GB | 10.9 GB | Too big | Get file → | |
| Q8_0 | 13.2 GB | 15.3 GB | Too big | Get file → | |
| BF16 | 20.5 GB | 23.5 GB | Too big | Get file → |
For 12 GB VRAM use the Q4_K quant (9.26 GB, ~10.9 GB memory at 8 k context). If 16 GB is available, Q8_0 (13.18 GB, ~15.3 GB memory) offers higher fidelity. bf16 (20.55 GB, ~23.5 GB memory) requires 24 GB VRAM but gives the best quality.
What GPU do you need for MOSS-TTS-v1.5?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q4_K (9.3 GB file, about 10.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (13.2 GB file, about 15.3 GB in use).
- 24 GB VRAM: the best quant that fits is BF16 (20.5 GB file, about 23.5 GB in use).
Which MOSS-TTS-v1.5 quant should you choose?
Q8_0 (13.2 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.
If memory is tight, the smallest file is Q4_K at 9.3 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about MOSS-TTS-v1.5
How much VRAM does MOSS-TTS-v1.5 need?
About 15.3 GB at Q8_0 with an 8K context window. That is the 13.2 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 MOSS-TTS-v1.5 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of MOSS-TTS-v1.5 GGUF?
It depends on the quant: Q4_K is 9.3 GB, Q8_0 is 13.2 GB, BF16 is 20.5 GB.
Where can I download MOSS-TTS-v1.5 GGUF?
The files are on Hugging Face in the christopherthompson81/MOSS-TTS-v1.5-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 MOSS-TTS-v1.5 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 MOSS‑TTS‑v1.5 on 8 GB VRAM?
No. The smallest quant requires 9.26 GB file size and ~10.9 GB memory at 8 k context, exceeding 8 GB VRAM.
Is the model lossless?
The Q4_K quant is not lossless on wording; Q8_0 is exact on all tested sentences.
What GPU is needed for bf16?
A recent audio.cpp build on a GPU with at least 24 GB VRAM is required for bf16.