Gemini APIs reportedly understand PDFs with images, charts and handwriting
A user highlights a FinChartBench example and says Gemini 3.8 Flash can reverse engineer and recreate plots embedded in PDFs.
TLDR
A user says the Gemini APIs in Google AI Studio natively support PDF understanding across multiple languages, including documents with embedded images, charts and handwriting. They describe the process as uploading a file and prompting. In a follow-up reply, they say Gemini 3.8 Flash can also reverse engineer and recreate embedded plots, pointing to a FinChartBench example.
Combined views
8.7K
3 Sources, first seen ago
