Yes, DeepSeek can read PDFs: DeepSeek’s API documentation says you can upload a PDF and have the model answer questions about it, “like uploading files in DeepSeek Chat.”
As of November 15, 2026, according to DeepSeek’s API documentation:
- DeepSeek App and web: file input is already part of the chat experience, and on November 13, 2026, DeepSeek opened it to developers, per the change log.
- Two parsing modes: text_recognition gives the model text only, while rich_layout_recognition also passes the charts, photos and diagrams cropped from the document.
- API limits: PDFs up to 100 MiB and 1,000 pages, used with the deepseek-flash model. One request can reference up to 20 documents with at most 8 MiB of text in total.
- Cost: parsing is billed once at $0.225 per 1,000 pages, per the pricing page; each chat that references the file is billed at the model’s input token price.
- Not supported: password-protected PDFs.
Related: DeepSeek Application APIs: Power-User Guide, Can DeepSeek read images? and Can Gemini read PDFs?
Source: DeepSeek API Docs, File Parsing, DeepSeek API Docs, Change Log and DeepSeek API Docs, Models & Pricing
Image: DeepSeek logo by DeepSeek, MIT license, via Wikimedia Commons. Placed on a white background and resized. DeepSeek is a trademark of its owner. FreeAnswers is not affiliated with or endorsed by DeepSeek.