Downloads · 30 days
13
21% of all-time downloads
bschooled/caeleste-speech-moss-tts-v1.5-nf4
caeleste-speech-moss-tts-v1.5-nf4 is a machine learning model from bschooled. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
4-bit NF4 (bitsandbytes) quantization of OpenMOSS-Team/MOSS-TTS-v1.5.
Downloads · 30 days
13
21% of all-time downloads
All-time downloads
61
Public
Parameters
8.5B
6.7 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors6.7 GB · 100%
How the weights are stored.
U86.9B · 82%
From the Hugging Face model README
4-bit NF4 (bitsandbytes) quantization of
OpenMOSS-Team/MOSS-TTS-v1.5.
| Upstream repository | OpenMOSS-Team/MOSS-TTS-v1.5 |
| Upstream revision | cdd3b911b1585e3f2dbc7775ef10f9926f58850a |
| Upstream license | apache-2.0 |
| Quantization | bitsandbytes NF4, double quantization, bf16 compute |
| Original size | 15.83 GiB |
| Quantized size | 6.23 GiB |
Quality text-to-speech (Qwen3-8B backbone). This is the model that cannot fit a 16GB GPU unquantized.
The weight tensors have been replaced with their NF4-quantized equivalents and a
quantization_config block was added to config.json. Tokenizer, processor,
and any remote-code files are copied unmodified from the upstream revision above.
No architecture, vocabulary, or generation defaults were altered.
Loading requires bitsandbytes and a supported accelerator. NF4 kernels are
available for NVIDIA sm_75+ (Turing and newer) and for AMD ROCm on RDNA3 /
RDNA3.5 / CDNA.
from transformers import AutoModel
model = AutoModel.from_pretrained(
"bschooled/caeleste-speech",
subfolder="moss-tts-v1.5-nf4",
device_map={"": "cuda:0"},
trust_remote_code=True,
)
This derivative is distributed under the upstream license, apache-2.0.
All upstream copyright, patent, trademark and attribution notices are retained.
Refer to the upstream repository for the authoritative license text.