Join us October 22nd to hear Coglate-Palmolive, IDC, and Sequoia Capital discuss moving to a digital-first environment
Learn more


DocQuery, Impira's new open-source machine learning model, lets you ask questions about the data in semi-structured (like invoices) and unstructured documents (like contracts) using Large Language Models (LLMs).

Purchase order with chat bubbles in front. First bubble asks "What is the PO Number?" The second bubble says "05."

How to use DocQuery

Simply upload an image or PDF, ask a question (like, "What is the invoice number?"), and click Submit. You can also choose from the examples and ask questions about those documents.

DocQuery uses LayoutLMv1 fine-tuned on DocVQA, a document visual question answering dataset, as well as SQuAD, which boosts its English-language comprehension. DocQuery is MIT-licensed and available on Github.

Try our demo: