7h agoKeshigeyan Chandrasegaran and Kyle Sargent release GPIC, a fully permissive 100 million-image dataset for training visual modelsThe corpus spans 28 trillion pixels of visual dataSentimentSentimentPos100%Neg0%Users praise the GPIC 100M permissive image-text dataset release as a valuable open contribution for training visual AI models.1 comment with sentiment. View comments.