Introducing HAMMR: hierarchical multimodal agents that handle a broad range of VQA tasks within a single system (counting, spatial reasoning, OCR, visual pointing, external knowledge, and more). arxiv.org/abs/2404.05465 Lluís Castrejón @tejmensink Howard Zhou André Araujo Jasper Uijlings

Today, we’re announcing Veo 2: our state-of-the-art video generation model which produces realistic, high-quality clips from text or image prompts. 🎥 We’re also releasing an improved version of our text-to-image model, Imagen 3 - available to use in ImageFX through

Breaking news from Text-to-Image Arena! 🖼️✨ Google DeepMind’s Imagen 3 debuts at #1, surpassing Recraft-v3 with a remarkable +70-point lead! Congrats to the Google Imagen team for setting a new bar! Try the best text2image at LMArena and cast your vote! More analysis👇