OM

OmniVoice-tokenizer

OmnivoiceGGUF by ServeurpersoUpdated 2026-09-09
Download on Hugging Face
Parameters613M
Context—
Layers—
Architectureomnivoice-tokenizer
Smallest quant0.3 GB
Quick answer

OmniVoice-tokenizer needs about 0.8 GB of memory at Q4_K_M (a 0.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 quants, from 0.3 GB (Q4_K_M) to 2.5 GB (F32).

OmniVoice-tokenizer: 0.613B multilingual TTS tokenizer

OmniVoice-tokenizer is a 0.613B parameter tokenizer for the OmniVoice multilingual zero‑shot text‑to‑speech system. It supports over 600 languages and is part of the OmniVoice family. The model is released under CC‑BY‑NC‑4.0.

Good at

  • Multilingual TTS tokenization
  • Zero‑shot voice cloning
  • Fine‑grained control of speech output

Limits

  • No vision support
  • Requires paired base model
  • License is non‑commercial

Best for

  • Voice cloning from short audio
  • Voice design with speaker attributes
  • Generating speech in rare languages

Who it’s for. Developers building multilingual TTS applications who need a lightweight tokenizer that works with OmniVoice. It is suitable for CPU, CUDA, ROCm, Metal, and Vulkan backends via omnivoice.cpp.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M0.3 GB 0.8 GB
Fits Get file →
Q8_00.3 GB 0.8 GB
Fits Get file →
BF160.4 GB 0.9 GB
Fits Get file →
Q4_K_M0.4 GB 1.0 GB
Fits Get file →
Q8_00.7 GB 1.2 GB
Fits Get file →
F320.7 GB 1.3 GB
Fits Get file →
BF161.2 GB 1.9 GB
Fits Get file →
F322.5 GB 3.3 GB
Fits Get file →

For 8 GB VRAM, the Q4_K_M 0.25 GB file uses about 0.8 GB memory at 8 k context, which is safe. If more precision is needed, the Q8_0 0.29 GB file also fits within the same budget. Larger models such as F32 2.46 GB require more than 8 GB VRAM.

What GPU do you need for OmniVoice-tokenizer?

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 F32 (2.5 GB file, about 3.3 GB in use).

Which OmniVoice-tokenizer quant should you choose?

Q4_K_M is the usual starting point for OmniVoice-tokenizer: it balances file size and quality well. Here it is a 0.3 GB file.

The Q8_0 file is 0.3 GB, about 1.1 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureomnivoice-tokenizer
Parameters613M

Source

Quantized byServeurperso
FamilyOmnivoice
Downloads (HF)44,796
Last updated2026-09-09

Frequently asked questions about OmniVoice-tokenizer

How much VRAM does OmniVoice-tokenizer need?

About 0.8 GB at Q4_K_M with an 8K context window. That is the 0.3 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 OmniVoice-tokenizer on an 8 GB GPU?

Yes, with F32: the 2.5 GB file needs about 3.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of OmniVoice-tokenizer GGUF?

It depends on the quant: Q4_K_M is 0.3 GB, Q8_0 is 0.3 GB, BF16 is 0.4 GB, Q4_K_M is 0.4 GB, Q8_0 is 0.7 GB, F32 is 0.7 GB and 2 more.

Where can I download OmniVoice-tokenizer GGUF?

The files are on Hugging Face in the Serveurperso/OmniVoice-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 OmniVoice-tokenizer use?

Hugging Face lists the license as cc-by-nc-4.0. 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 OmniVoice-tokenizer on 8 GB VRAM?

Yes, the Q4_K_M 0.25 GB quant uses about 0.8 GB memory at 8 k context, well within 8 GB VRAM.

What license applies to OmniVoice-tokenizer?

The tokenizer is licensed under CC‑BY‑NC‑4.0, meaning it is non‑commercial.

Does this tokenizer support vision tasks?

No, it is a text‑to‑speech tokenizer and has no vision capabilities.

Other GGUFs of OmniVoice

Similar models

Browse all →

More from Serveurperso

All by Serveurperso →

More Omnivoice models

All Omnivoice →

Other tiny models

All tiny models →