Orpheus 3b Korean FT Q8 0.gguf

🚀 Orpheus-3b-FT-Q8_0

Orpheus-3b-FT-Q8_0 is a quantized Text-to-Speech model that efficiently converts text into high - quality natural speech, supporting multiple voices and emotional expressions.

🚀 Quick Start

Download the Model

1. Download this quantised model from lex - au's Orpheus - FASTAPI collection.

Load the Model

2. Load the model in your preferred inference server and start the server. The compatible inference servers include:

• GPUStack - GPU optimised LLM inference server (My pick) - supports LAN/WAN tensor split parallelisation
• LM Studio - Load the GGUF model and start the local server
• llama.cpp server - Run with the appropriate model parameters
• Any compatible OpenAI API - compatible server

Set up the Front - end

3. Clone the Orpheus - FastAPI repository:

git clone https://github.com/Lex-au/Orpheus-FastAPI.git
cd Orpheus-FastAPI

4. Configure the FastAPI server to connect to your inference server by setting the ORPHEUS_API_URL environment variable.

5. Follow the complete installation and setup instructions in the repository README.

✨ Features

• Multiple Voice Options: Offers 2 distinct voice options with different characteristics.
• Emotion Tag Support: Supports emotion tags like laughter, sighs, etc., to add expressiveness to speech.
• CUDA Acceleration: Optimised for CUDA acceleration on RTX GPUs.
• High - Quality Audio: Produces high - quality 24kHz mono audio.
• Conversational Naturalness: Fine - tuned for conversational naturalness.

📦 Installation

This model is designed to be used with an LLM inference server that connects to the [Orpheus - FastAPI](https://github.com/Lex - au/Orpheus - FastAPI) frontend. You can load the quantised model into any of the following LLM inference servers:

• GPUStack
• LM Studio
• llama.cpp server
• Any compatible OpenAI API - compatible server

💻 Usage Examples

Available Voices

The model supports 2 different voices:

• 유나: Female, Korean, melodic
• 준서: Male, Korean, confident

Emotion Tags

You can add expressiveness to speech by inserting tags:

• <laugh>, <chuckle>: For laughter sounds
• <sigh>: For sighing sounds
• <cough>, <sniffle>: For subtle interruptions
• <groan>, <yawn>, <gasp>: For additional emotional expression

📚 Documentation

Model Description

Orpheus - 3b - FT - Q8_0 is a 3 billion parameter Text - to - Speech model. It converts text inputs into natural - sounding speech, supporting multiple voices and emotional expressions. The model has been quantised to 8 - bit (Q8_0) format for efficient inference, making it accessible on consumer hardware.

🔧 Technical Details

🚫 Limitations

• Best performance achieved on CUDA - compatible GPUs.
• Generation speed depends on GPU capability.

📄 License

This model is available under the [Apache License 2.0](https://www.apache.org/licenses/LICENSE - 2.0).

Citation & Attribution

The original Orpheus model was created by Canopy Labs. This repository contains a quantised version optimised for use with the Orpheus - FastAPI server.

If you use this quantised model in your research or applications, please cite:

@misc{orpheus-tts-2025,
  author = {Canopy Labs},
  title = {Orpheus-3b-0.1-ft: Text-to-Speech Model},
  year = {2025},
  publisher = {HuggingFace},
  howpublished = {\url{https://huggingface.co/canopylabs/orpheus-3b-0.1-ft}}
}

@misc{orpheus-quantised-2025,
  author = {Lex-au},
  title = {Orpheus-3b-FT-Q8_0: Quantised TTS Model with FastAPI Server},
  note = {GGUF quantisation of canopylabs/orpheus-3b-0.1-ft},
  year = {2025},
  publisher = {HuggingFace},
  howpublished = {\url{https://huggingface.co/lex-au/Orpheus-3b-FT-Q8_0.gguf}}
}