AI startup Underdog has released a browser-based game to demonstrate its new text-to-speech model, Woof, which weighs in at just 31.7MB and 31 million parameters. The demo went live on November 5, 2026, and the model is available for anyone to download.
The game is set in a retro-themed room featuring a typewriter, a CRT monitor, and a desk lamp. Players can type anything using their keyboard and the model will read it back aloud. A music tuner in the game lets users rotate a dial to generate different styles of music. An in-game warning about “intrusive” cloud providers appears to be a reference to cloud-based AI models.
According to Underdog, Woof can run on most devices without an internet connection, making it an example of what the industry calls an edge model — software designed to process data locally rather than on remote servers.
Underdog founder Sigil Wen told TechCrunch that the project grew out of a desire to build AI tools that keep data in users’ hands. “I think we will be able to run most capable models of today on the devices you already own within months. We just want to train the best open models that are American-made, that are open, that are capable and allow people to have private AI,” Wen said.
The Woof release follows Underdog’s launch of Bark, a 167MB speech-to-text model also built for edge devices, which the company released weeks earlier. Underdog has additionally released a 27-billion-parameter model designed to run on laptops, along with a waitlisted app positioned as a locally hosted alternative to Muse and Instinct.
A model of Woof’s size is unlikely to replace larger AI systems, but it could support local voice assistant experiences or personalized podcast generation, the source material suggests.
Source: TechCrunch