Llasa-3B
Update （2025-05-10): Sometimes I find that top_p=0.95 and temperature=0.9 produce more stable results.
text → audio · made by HKUST Audio
Sold by
Nobody in the catalogue publishes a price for this yet.
About
Llasa-3B — an audio model from HKUST Audio.
It takes text and returns audio. It was published in January 2025. The catalogue files it under speak. Its weights are published, but the licence forbids selling what it produces.
Every current figure
- Maker
- HKUST Audio
- Register
- model
- Takes
- text
- Returns
- audio
- Published
- January 2025
- Parameters
- 4 billion · read from its own weights
- Licence
- cc-by-nc-4.0
Known as 1 name
HKUSTAudio/Llasa-3B