Model releaseinfo
Underdog demos 32MB Woof speech model for edge devices
Original title: Underdog shows off a 32MB speech model in new demo
AI summary · x-ai/grok-4.7Summaries by Vani AI
The 32MB Woof model runs locally in browsers and supports typing to speech without internet.
Underdog released a 32MB speech model called Woof for a retro-themed game demo. The model has 31 million parameters and works entirely offline in the browser. Founder Sigil Wen highlighted plans for more capable local models and an open-source American AI stack.
Key facts
- 31 million parameters in 32MB Woof model
- Runs on most devices without internet
- Open model download available
- Supports typewriter and music tuner features
- Launched alongside 167MB Bark model
- Vendors
- Underdog
From the original article
2 min read at techcrunch.com
Despite its small size, Underdog's new model can generate both speech and music.
By Ivan Mehta · techcrunch.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.