Yes, DeepSeek can read images: its current model understands photos, screenshots, charts and scanned pages, and can describe them, read text from them and answer questions about them. Its replies are text only.
As of November 14, 2026, according to DeepSeek’s official API documentation:
- Model: DeepSeek-V4.1-Flash, released September 10, 2026, has native multimodal visual understanding, per the change log. In the API it is called deepseek-flash.
- Formats: JPEG, PNG, GIF and WebP.
- API limits: up to 600 images per request, 32 MiB per image sent inline or by URL (64 MiB through the Files API) and 8192 pixels per side. Each image uses at most 1,024 tokens.
- Documents: you can upload PDFs, Word and PowerPoint files and images for parsing, “like uploading files in DeepSeek Chat.” A standalone OCR API, added November 13, 2026, returns text only.
Related: DeepSeek V4.1 Flash: Power-User Guide, DeepSeek Application APIs: Power-User Guide and Can Claude read images?
Source: DeepSeek API Docs, Vision, DeepSeek API Docs, Change Log and DeepSeek API Docs, File Parsing
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.