KI

kitten tts2

KittenGGUF by dignomeUpdated 2026-10-23
Download on Hugging Face
Parameters—
Context—
Layers—
Architectureaudiocpp
Smallest quant1.4 GB
Quick answer

kitten tts2 needs about 3.0 GB of memory at its main file (a 2.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 quants, from 1.4 GB to 4.6 GB (BF16).

Kitten TTS 2 – multilingual text‑to‑speech model in GGUF format

Kitten TTS 2 is a 1.7B speech language model that generates S3 codec tokens for 24 kHz audio. It supports 115 voices across 16 languages and includes expression controls such as emotion tags and emphasis markers. The model is distributed as GGUF files with various quantizations.

Good at

  • Multilingual voice cloning
  • Emotion and emphasis control
  • Compact GGUF packaging

Limits

  • No vision support
  • Requires audio.cpp build with updated Kitten TTS 2 support
  • License is other, not standard

Best for

  • Developing TTS applications
  • Creating multilingual audio content
  • Testing voice cloning features

Who it’s for. Engineers and researchers building or experimenting with text‑to‑speech systems. The model is suitable for developers who need a self‑contained GGUF export and can run audio.cpp without Python or LibTorch.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.4 GB 2.1 GB
Fits Get file →
—1.4 GB 2.1 GB
Fits Get file →
—1.9 GB 2.6 GB
Fits Get file →
—2.3 GB 3.0 GB
Fits Get file →
—2.3 GB 3.0 GB
Fits Get file →
—3.3 GB 4.2 GB
Fits Get file →
—3.3 GB 4.2 GB
Fits Get file →
BF164.6 GB 5.7 GB
Fits Get file →

For 8 GB VRAM, use the 1.41 GB or 1.44 GB GGUF files; they need about 2.1 GB RAM at 8 k context. If you have 12 GB VRAM, the 2.27 GB files are safe, requiring roughly 3 GB RAM. For larger memory budgets, the 4.6 GB BF16 file uses about 5.7 GB RAM.

What GPU do you need for kitten tts2?

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 (2.3 GB file, about 3.0 GB in use).
  • 6 GB VRAM: the best quant that fits is BF16 (4.6 GB file, about 5.7 GB in use).

Which kitten tts2 quant should you choose?

If memory is tight, the smallest file is the first one in the table at 1.4 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureaudiocpp

Source

Quantized bydignome
FamilyKitten
Licenseother
Downloads (HF)67,133
Last updated2026-10-23

Frequently asked questions about kitten tts2

How much VRAM does kitten tts2 need?

About 3.0 GB at its main file with an 8K context window. That is the 2.3 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 kitten tts2 on an 8 GB GPU?

Yes, with BF16: the 4.6 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of kitten tts2 GGUF?

It depends on the quant: file is 1.4 GB, file is 1.4 GB, file is 1.9 GB, file is 2.3 GB, file is 2.3 GB, file is 3.3 GB and 2 more.

Where can I download kitten tts2 GGUF?

The files are on Hugging Face in the dignome/kitten_tts2 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does kitten tts2 use?

Hugging Face lists the license as other. 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 Kitten TTS 2 on 8 GB VRAM?

Yes, the 1.41 GB and 1.44 GB GGUF files fit within 8 GB VRAM and need about 2.1 GB RAM at 8 k context.

Does the model support non‑English text?

Non‑English text can be processed, but the normalizer is English‑tuned; use normalize=False to avoid mangling numbers and dates.

What quantizations are available?

The model ships in unknown precision GGUF files and a BF16 file. Q8 is the default; Q4 is experimental and requires a compatible audio.cpp build.

More from dignome

All by dignome →