@ajitomatix tests TypeSafe Jev 1.14 experimental image support on SOS Draw, comparing text, image, and hybrid against a handcrafted stroke-to-text pipeline.

App post with a photo. When TypeSafe AI released Jev 1.14 with experimental image support, @ajitomatix tested it on SOS Draw. The question in the post: can TypeSafe recognize a player's raw canvas drawing better than their handcrafted stroke-to-text pipeline? They ran a side-by-side benchmark of text, image, and hybrid. The post does not give the scores. For people trying Jev image input on drawings.