Experimental multilingual
What MOSS-TTS-Nano needs to run
MOSS-TTS-Nano is a 100-million-parameter multilingual model with an official ONNX browser stack. It supports short voice references, 20 listed languages, streaming-oriented generation, and 48 kHz stereo output, with a larger download than the everyday models.
Experimental browser integration
MOSS-TTS-Nano is not ready in the generator yet. Review its size, runtime needs, and current limitations below, or use one of the available models . The compatibility check can also show whether your browser exposes the hardware features this model would need.
Specifications
| Developer | OpenMOSS / MOSI.AI (Fudan) |
|---|---|
| Parameters | 100 million |
| Download size | 684 MB (fp32 (official ONNX export)) |
| Output | 48.0 kHz stereo |
| Listed voices | 17 |
| Languages | Chinese, English, German, Spanish, French, Japanese, Italian, Hungarian, Korean, Russian, Persian, Arabic, Polish, Portuguese, Czech, Danish, Swedish, Greek, Turkish, Hebrew |
| Voice cloning | Yes |
| Streaming | Yes — audio starts before generation finishes |
| License | Apache-2.0 |
| Speed | Uses the project's split browser graphs and may take several seconds per sentence on a CPU. |
| Runtime | WebAssembly (CPU) |
Specifications checked 2026-07-21 against:
Explore related downloads
The browser studio and downloadable catalog are separate. Use the catalog to find independently published model files and check each source before downloading.
Good fit when
- +Official onnxruntime-web deployment files
- +Voice references across 20 listed languages
- +48 kHz stereo output
Before you choose it
- −Approximately 684 MB across the model and audio tokenizer
- −Slower than the lightweight CPU models
- −Only the included and appropriately licensed reference voices are shown
Voices (19)
Questions about MOSS-TTS-Nano
Does AIVoices charge to use MOSS-TTS-Nano?
AIVoices does not charge generation credits for this browser model. The listed model license is Apache-2.0. Check the upstream model and voice terms before use.
What hardware does MOSS-TTS-Nano need?
A current desktop browser and CPU are enough for this model. The 684 MB model is stored in your browser after the first download.
Is my text private when using MOSS-TTS-Nano here?
The supported model executes inside your browser tab. AIVoices does not send the text to a server for speech generation.
Can I use MOSS-TTS-Nano output commercially?
Commercial use depends on the model, selected voice, and source-data terms. This page lists Apache-2.0 for the model, but you should still review the current upstream terms.