X-citedOpen sourceUC-148classification
NotHotDog multimodal decision app
The open-source nothotdog demo uses a vision model to write one sentence about an image and Jev to decide from that description; the author reports 1002 ms end-to-end versus 327 ms for the Jev half and displays both bars.
NotHotDog multimodal decision app
The open-source nothotdog demo uses a vision model to write one sentence about an image and Jev to decide from that description; the author reports 1002 ms end-to-end versus 327 ms for the Jev half and displays both bars.
Category: classification
Primitives: choice, noul
Source type: x
Open source: yes
Date: 2026-09-19
Citations
- https://x.com/iamAnish/status/2101306035525107826
- https://github.com/anishsrinivasan/nothotdog