Instructions to use Aratako/MioTTS-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Aratako/MioTTS-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Aratako/MioTTS-GGUF",
	filename="MioTTS-0.1B-BF16.gguf",
)

llm.create_chat_completion(
	messages = "\"The answer to the universe is 42\""
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Aratako/MioTTS-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Aratako/MioTTS-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Aratako/MioTTS-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Aratako/MioTTS-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Aratako/MioTTS-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Aratako/MioTTS-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Aratako/MioTTS-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Aratako/MioTTS-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Aratako/MioTTS-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Aratako/MioTTS-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use Aratako/MioTTS-GGUF with Ollama:
```
ollama run hf.co/Aratako/MioTTS-GGUF:Q4_K_M
```

Unsloth Studio new

How to use Aratako/MioTTS-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Aratako/MioTTS-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Aratako/MioTTS-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Aratako/MioTTS-GGUF to start chatting

Docker Model Runner
How to use Aratako/MioTTS-GGUF with Docker Model Runner:
```
docker model run hf.co/Aratako/MioTTS-GGUF:Q4_K_M
```

Lemonade

How to use Aratako/MioTTS-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Aratako/MioTTS-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.MioTTS-GGUF-Q4_K_M

List all available models

lemonade list

MioTTS-GGUF

This repository contains GGUF quantized versions of the MioTTS models. MioTTS is a lightweight, high-speed Text-to-Speech (TTS) model family designed for high-quality English and Japanese speech generation.

For model details, usage, and citations, please refer to the original model cards (linked below).

📦 Available Models & Files

Model Size	Quantization	File Name	Size	Original Model
0.1B	BF16	`MioTTS-0.1B-BF16.gguf`	232 MB	Link
	Q8_0	`MioTTS-0.1B-Q8_0.gguf`	125 MB
	Q6_K	`MioTTS-0.1B-Q6_K.gguf`	97.3 MB
	Q4_K_M	`MioTTS-0.1B-Q4_K_M.gguf`	79.6 MB
0.4B	BF16	`MioTTS-0.4B-BF16.gguf`	736 MB	Link
	Q8_0	`MioTTS-0.4B-Q8_0.gguf`	392 MB
	Q6_K	`MioTTS-0.4B-Q6_K.gguf`	304 MB
	Q4_K_M	`MioTTS-0.4B-Q4_K_M.gguf`	239 MB
0.6B	BF16	`MioTTS-0.6B-BF16.gguf`	1.22 GB	Link
	Q8_0	`MioTTS-0.6B-Q8_0.gguf`	653 MB
	Q6_K	`MioTTS-0.6B-Q6_K.gguf`	506 MB
	Q4_K_M	`MioTTS-0.6B-Q4_K_M.gguf`	408 MB
1.2B	BF16	`MioTTS-1.2B-BF16.gguf`	2.39 GB	Link
	Q8_0	`MioTTS-1.2B-Q8_0.gguf`	1.27 GB
	Q6_K	`MioTTS-1.2B-Q6_K.gguf`	983 MB
	Q4_K_M	`MioTTS-1.2B-Q4_K_M.gguf`	751 MB
1.7B	BF16	`MioTTS-1.7B-BF16.gguf`	3.5 GB	Link
	Q8_0	`MioTTS-1.7B-Q8_0.gguf`	1.86 GB
	Q6_K	`MioTTS-1.7B-Q6_K.gguf`	1.44 GB
	Q4_K_M	`MioTTS-1.7B-Q4_K_M.gguf`	1.13 GB
2.6B	BF16	`MioTTS-2.6B-BF16.gguf`	5.19 GB	Link
	Q8_0	`MioTTS-2.6B-Q8_0.gguf`	2.76 GB
	Q6_K	`MioTTS-2.6B-Q6_K.gguf`	2.13 GB
	Q4_K_M	`MioTTS-2.6B-Q4_K_M.gguf`	1.58 GB

🚀 Usage

Please check the official inference repository for instructions on how to run these models.

👉 GitHub: Aratako/MioTTS-Inference

📜 License

Please note that the license differs depending on the model size (inherited from their respective base models). Please check the original model card for the specific license terms before use.

0.1B: Falcon-LLM License
0.4B, 1.2B, 2.6B: LFM Open License v1.0
0.6B, 1.7B: Apache 2.0

Downloads last month: 692

GGUF

Model size

0.1B params

Architecture

falcon-h1

Hardware compatibility

4-bit

6-bit

8-bit

16-bit

Collection including Aratako/MioTTS-GGUF

MioTTS

Collection

12 items • Updated Feb 14 • 10