PaliGemma - Google Vision-Language Model

Source: https://huggingface.co/google/paligemma2-3b-ft-docci-448 License: gemma Type: Vision-Language Model (VLM)

Installation

pip install transformers accelerate pillow

Usage

from transformers import PaliGemmaProcessor, PaliGemmaForConditionalGeneration
model = PaliGemmaForConditionalGeneration.from_pretrained("google/paligemma2-3b-ft-docci-448")

Model Weights

Downloaded automatically via HuggingFace Hub on first use. Cached at: ~/.cache/huggingface/hub/models--google--paligemma2-3b-ft-docci-448/