Search for a command to run...
olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models