PaliGemma 2 3B Mix 448

Parameters

2.9B

Published

Feb 19, 2025

PaliGemma 2 3B Mix 448 is an AI model developed by Google (United States), first published in February 2025. It works in the vision, multimodal and language domain, on tasks such as image captioning, visual question answering, object detection and 3 more.

Epoch AI has no training-compute estimate for this model. The model has 2,920,000,000 parameters. It was trained on roughly 51.2B datapoints. Training ran on 256 Google TPU v5e.

Access: Open weights (non-commercial). Its weights are openly available. It is built on top of PaliGemma 2 3B Mix 224. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Google
Country of organization: United States
Domain: Vision, Multimodal, Language
Task: Image captioning, Visual question answering, Object detection, Object recognition, Image segmentation, Character recognition (OCR)
Parameters: 2,920,000,000
Dataset size: 51.2B
Training hardware: Google TPU v5e
Chips used: 256
Training power draw: 113.1 kW
Model accessibility: Open weights (non-commercial)
Open weights: Yes
Base model: PaliGemma 2 3B Mix 224
Epoch confidence: Confident