Text-to-Audio
MLX
Safetensors
Chinese
English
moss_tts_delay
sound-effect
moss
custom_code
4-bit precision
Instructions to use dpetrusevich/MOSS-SoundEffect-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use dpetrusevich/MOSS-SoundEffect-MLX-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir MOSS-SoundEffect-MLX-4bit dpetrusevich/MOSS-SoundEffect-MLX-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
MOSS-SoundEffect — MLX 4-bit (Swift-ready re-host)
MLX 4-bit MOSS-SoundEffect (MossTTSDelay arch) for on-device text→sound-effect
generation in the Trails app (macOS, via mlx-audio-swift's MossTTSModel). The
audio tokenizer is bundled under audio_tokenizer/ so the Swift loader finds
it in one snapshot (no separate codec download).
Attribution (all Apache-2.0)
- Base:
OpenMOSS-Team/MOSS-SoundEffect - MLX 4-bit model:
mlx-community/MOSS-SoundEffect-MLX-4bit - MLX audio tokenizer:
appautomaton/openmoss-audio-tokenizer-mlx
Re-hosted unmodified (model at root + tokenizer under audio_tokenizer/).
- Downloads last month
- -
Model size
1B params
Tensor type
BF16
·
U32 ·
Hardware compatibility
Log In to add your hardware
4-bit
Model tree for dpetrusevich/MOSS-SoundEffect-MLX-4bit
Base model
OpenMOSS-Team/MOSS-SoundEffect