Post
34
d1-2B on a Galaxy S26: four answers about one photo, boxes included, in 1,983 ms.
Liquid AI's decision model for text, pictures and video, converted to LiteRT: one forward pass, no tokens generated, typed answers (yes or no, a choice, a number with an interval, a value with its span, boxes) with their probabilities.
In the clip, a photo of a bicycle in snow (CC0) and four questions. One tap: every wheel, 2 boxes (found 1.00 and 0.89); the saddle, one box (confidence 0.94); how many bicycles, 1 (80% between 1 and 1); a car, No (P(yes) 0.08). Answered in 1,983 ms on this phone, 423 tokens on the 512-token graph, S26 GPU through LiteRT.
To run it: the repo has the LiteRT graphs (five also compiled for the S26's NPU) and a Python host for desktops, no PyTorch. The app in the clip is my Kotlin sample: add a photo or paste a message, write your own questions. No internet permission.
https://github.com/john-rocky/LiteRT-Models/tree/main/d1-2b
litert-community/d1-2B-LiteRT
Liquid AI's decision model for text, pictures and video, converted to LiteRT: one forward pass, no tokens generated, typed answers (yes or no, a choice, a number with an interval, a value with its span, boxes) with their probabilities.
In the clip, a photo of a bicycle in snow (CC0) and four questions. One tap: every wheel, 2 boxes (found 1.00 and 0.89); the saddle, one box (confidence 0.94); how many bicycles, 1 (80% between 1 and 1); a car, No (P(yes) 0.08). Answered in 1,983 ms on this phone, 423 tokens on the 512-token graph, S26 GPU through LiteRT.
To run it: the repo has the LiteRT graphs (five also compiled for the S26's NPU) and a Python host for desktops, no PyTorch. The app in the clip is my Kotlin sample: add a photo or paste a message, write your own questions. No internet permission.
https://github.com/john-rocky/LiteRT-Models/tree/main/d1-2b
litert-community/d1-2B-LiteRT