lucataco/deepseek-ocr

Convert documents to markdown, extract raw text, and locate specific content

Capabilities

Reference Images

Community model (estimated from hardware time)

Name	Type	Description	Default	Constraints
`image`*	string(uri)	Input image to perform OCR on (supports documents, charts, tables, etc.)	`—`	—
`reference_text`	string	Reference text to locate in the image (only used for 'Locate Object by Reference' task). Example: 'the teacher', '20-10', 'a red car'	`""`	—
`resolution_size`	string	Model resolution size - affects speed and accuracy trade-off	`"Gundam (Recommended)"`	Gundam (Recommended)TinySmallBaseLarge
`task_type`	string	Type of OCR task to perform	`"Convert to Markdown"`	Convert to MarkdownFree OCRParse FigureLocate Object by Reference

imagerequiredstring

Input image to perform OCR on (supports documents, charts, tables, etc.)

reference_textstring

Reference text to locate in the image (only used for 'Locate Object by Reference' task). Example: 'the teacher', '20-10', 'a red car'

Default: ""

resolution_sizestring

Model resolution size - affects speed and accuracy trade-off

Default: "Gundam (Recommended)"

Gundam (Recommended)TinySmallBaseLarge

task_typestring

Type of OCR task to perform

Default: "Convert to Markdown"

Convert to MarkdownFree OCRParse FigureLocate Object by Reference

Version: cb3b474fbfc5Updated: 7/25/202652.4K runs