Server setup required

What MOSS-TTS (8B) needs to run

MOSS-TTS is the 8-billion-parameter flagship in the MOSS speech family. Its model size, memory requirements, and lack of a browser export make it a server deployment rather than an on-device option. Use MOSS-TTS-Nano here for the browser-oriented member of the family.

Separate server setup required

This model is too large for the current browser runtimes and does not have a working browser export. For a model from the same family that is designed for browser use, try MOSS-TTS-Nano. The full model requires a separate server-class installation.

Specifications

DeveloperOpenMOSS / MOSI.AI (Fudan)
Parameters8000 million
Download sizeabout 4.1 GB for the listed Q8 GGUF, plus tokenizer files (Q8 GGUF (server-side))
Output24.0 kHz mono
Listed voices1
LanguagesChinese, Cantonese, English, Arabic, Czech, Danish, German, Dutch, and more (v1.5 multilingual)
Voice cloningYes
StreamingYes — audio starts before generation finishes
LicenseApache-2.0
SpeedRequires server-class GPU resources and does not run in this browser tool.
RuntimeSeparate server or workstation setup

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

  • +Flagship model in the open MOSS speech family
  • +Voice references, multilingual generation, and streaming upstream
  • +Shares a model family with the browser-oriented Nano release

Before you choose it

  • Does not run in a browser
  • Requires a separate server-class setup
  • No browser ONNX export is available for the full model

Questions about MOSS-TTS (8B)

Does AIVoices charge to use MOSS-TTS (8B)?

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 (8B) need?

This model requires a separate server or workstation setup and does not have a working browser export.

Is my text private when using MOSS-TTS (8B) 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 (8B) 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.