https://www.infoq.com InfoQ PDF feed https://www.infoq.com/news/2025/09/finepdfs/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=PDF <img src="https://res.infoq.com/news/2025/09/finepdfs/en/headerimage/generatedHeaderImage-1757926150946.jpg"/><p>Hugging Face has unveiled FinePDFs, the largest publicly available corpus built entirely from PDFs. The dataset spans 475 million documents in 1,733 languages, totaling roughly 3 trillion tokens. At 3.65 terabytes in size, FinePDFs introduces a new dimension to open training datasets by tapping into a resource long considered too complex and expensive to process.</p> <i>By Robert Krzaczyński</i> PDF Artificial Intelligence Hugging Face Large language models AI, ML & Data Engineering news Mon, 15 Sep 2025 08:55:00 GMT https://www.infoq.com/news/2025/09/finepdfs/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=PDF Robert Krzaczyński 2025-09-15T08:55:00Z /news/2025/09/finepdfs/en