SN

snac 24khz

SnacGGUF by cstrUpdated 2026-08-02
Download on Hugging Face
Parameters13M
Context—
Layers—
Architecturecrispasr.reference
Smallest quant0.0 GB
Quick answer

snac 24khz needs about 0.5 GB of memory at its main file (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.0 GB to 0.0 GB.

SNAC 24 kHz audio codec for speech synthesis

SNAC compresses 24 kHz audio into discrete codes at 0.98 kbps. It uses 3 RVQ levels with token rates of 12, 23, and 47 Hz. The model is small (~25 MB) and licensed under MIT.

Good at

  • Speech synthesis codec
  • Low‑bitrate audio compression
  • Compatible with Orpheus‑TTS

Limits

  • No vision support
  • Single audio channel only
  • No quantisation available

Best for

  • Encoding speech for TTS
  • Reconstructing audio from tokens
  • Integrating with CrispASR

Who it’s for. Developers building TTS pipelines that require a lightweight codec. Researchers needing a small, open‑source audio codec for 24 kHz speech data.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 0.5 GB
Fits Get file →
—0.0 GB 0.5 GB
Fits Get file →

The two GGUF files are 0.01 GB and 0.03 GB. Both estimate 0.5 GB of VRAM at 8 k context. For 8 GB VRAM, either file will fit comfortably. If memory is very tight, choose the 0.01 GB file.

What GPU do you need for snac 24khz?

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 the main file (0.0 GB file, about 0.5 GB in use).

Specifications

Architecture from the GGUF header

Architecturecrispasr.reference
Parameters13M

Source

Quantized bycstr
FamilySnac
Licensemit
Downloads (HF)1,081
Last updated2026-08-02

Frequently asked questions about snac 24khz

How much VRAM does snac 24khz need?

About 0.5 GB at its main file with an 8K context window. That is the 0.0 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 snac 24khz on an 8 GB GPU?

Yes, with the smaller files: the 0.0 GB file needs about 0.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of snac 24khz GGUF?

It depends on the quant: file is 0.0 GB, file is 0.0 GB.

Where can I download snac 24khz GGUF?

The files are on Hugging Face in the cstr/snac-24khz-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 snac 24khz use?

Hugging Face lists the license as mit. 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 SNAC on 8 GB VRAM?

Yes. Both GGUF files estimate 0.5 GB VRAM at 8 k context, well below 8 GB.

Does SNAC support multi‑channel audio?

No. All models support only single audio channel (mono).

Is there a quantised version of SNAC?

No. The model is small enough that quantisation is not provided and is not recommended.

Other GGUFs of snac_24khz

Similar models

Browse all →

More from cstr

All by cstr →

More Snac models

All Snac →

Other tiny models

All tiny models →