Install Now
How to install OpenMOSS/MOSS-TTS. Official installation instructions. Project installation guide. Framework or library installation. Build installation instructions. How to setup and install from GitHub. Installation steps and setup instructions. Official docs and installation instructions GitHub.
OpenMOSS/MOSS-TTS Installation Guide
How to install OpenMOSS/MOSS-TTS. Official project installation instructions and setup guide.
⚡
Automated Install (Recommended)
Quick installation instructions for OpenMOSS/MOSS-TTS. This is the fastest way to complete project installation and setup.
Install via curl
curl -fsSL https://hexmos.com/ipm-install | bash &&
ipm i OpenMOSS/MOSS-TTSor
Install via npx
npx @hexmos/ipm i OpenMOSS/MOSS-TTSPrerequisites
Python
languageVersion: >=3.10
PyTorch
libraryVersion: >=2.0
CUDA
toolkitVersion: >=12.8
Conda
package managerVersion: >=23.1
uv
package managerVersion: >=0.1.14
llama.cpp
libraryVersion: >=0.9.0
onnxruntime-gpu
librarytensorrt
libraryManual Installation Methods
Manual installation instructions. How to install from GitHub source.
Environment Setup using Conda
conda create -n moss-tts python=3.12 -y
conda activate moss-tts
git clone https://github.com/OpenMOSS/MOSS-TTS.git
cd MOSS-TTS
pip install --extra-index-url https://download.pytorch.org/whl/cu128 -e ".[torch-runtime]"Environment Setup using uv
git clone https://github.com/OpenMOSS/MOSS-TTS.git
cd MOSS-TTS
uv venv --python 3.12 .venv
source .venv/bin/activate
uv pip install --torch-backend cu128 -e ".[torch-runtime]"Optional: Install FlashAttention 2 (using Conda/pip)
git clone https://github.com/OpenMOSS/MOSS-TTS.git
cd MOSS-TTS
pip install --extra-index-url https://download.pytorch.org/whl/cu128 -e "build-backend=setuptools.build_meta --config-settings=""config_settings.build_backend.cpp_args=["-fno-matmul]"" --config-settings=""config_settings.build_backend.common_args=["-fno-matmul]"" --config-settings=""config_settings.build_backend.cxx_args=["-fno-matmul]"" --no-build-isolation -e .[torch-runtime,flash-attn]"Optional: Install FlashAttention 2 (using uv)
git clone https://github.com/OpenMOSS/MOSS-TTS.git
cd MOSS-TTS
uv venv --python 3.12 .venv
source .venv/bin/activate
uv pip install --torch-backend cu128 -e ".[torch-runtime,flash-attn]"llama.cpp Backend (Torch-Free Inference)
pip install -e "build-backend=setuptools.build_meta --config-settings=""config_settings.build_backend.cpp_args=["-fno-matmul]"" --config-settings=""config_settings.build_backend.common_args=["-fno-matmul]"" --config-settings=""config_settings.build_backend.cxx_args=["-fno-matmul]"" --no-build-isolation -e .[llama-cpp-onnx]"
huggingface-cli download OpenMOSS-Team/MOSS-TTS-GGUF --local-dir weights/MOSS-TTS-GGUF
huggingface-cli download OpenMOSS-Team/MOSS-Audio-Tokenizer-ONNX --local-dir weights/MOSS-Audio-Tokenizer-ONNX
cd moss_tts_delay/llama_cpp && bash build_bridge.sh /path/to/llama.cpp && cd ../..
python -m moss_tts_delay.llama_cpp --config configs/llama_cpp/default.yaml --text "Hello, world!" --output output.wavSGLang Backend (Accelerated Inference)
git clone https://github.com/OpenMOSS/sglang.git
cd sglang && pip install -e ./python[all]
pip install nvidia-cudnn-cu12==9.16.0.29
huggingface-cli download OpenMOSS-Team/MOSS-TTS --local-dir weights/MOSS-TTS
huggingface-cli download OpenMOSS-Team/MOSS-Audio-Tokenizer --local-dir weights/MOSS-Audio-Tokenizer
python scripts/fuse_moss_tts_delay_with_codec.py --model-path weights/MOSS-TTS --codec-model-path weights/MOSS-Audio-Tokenizer --save-path weights/MOSS-TTS-Delay-With-Codec
sglang serve --model-path weights/MOSS-TTS-Delay-With-Codec --delay-pattern --trust-remote-codePost Installation Steps
- To run the basic usage example, ensure you have the necessary Python environment activated and the MOSS-TTS repository cloned and navigated into.
- The example Python script can be run directly after installation to generate audio samples.