PaliGemma2-3b-pt-448
PaliGemma 2 generates text from images and text inputs, excelling in multilingual visual-language tasks like image description and QA.
What is PaliGemma2-3b-pt-448
PaliGemma2-3b-pt-448 is a visual-language model developed by Google optimized for Portuguese with 3 billion parameters and trained on 448x448 pixel images. It excels in tasks like image description and visual question answering and supports efficient training on large datasets. Ideal for developers and researchers working with visual and textual data in Portuguese.