ComfyUI Node
DeepSeek OCR: Run
A ComfyUI node in DeepSeek/OCR with 8 inputs and 2 outputs.
DeepSeek OCR: Run
- model
- image
- text
- visualization
◄taskConvert to Markdown►
◄resolutionGundam►
◄output_typeall►
◄reference_text►
◄box_colorred►
◄box_width2►
CategoryDeepSeek/OCR
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| model | DPSK_OCR | — | |
| image | IMAGE | — | |
| task | COMBO | Convert to Markdown | 4 options: Free OCR, Convert to Markdown, Parse Figure, Locate by Reference |
| resolution | COMBO | Gundam | 5 options: Gundam, Tiny, Small, Base, Large |
| output_type | COMBO | all | 3 options: text, image, all |
| reference_textopt | STRING | — | |
| box_coloropt | STRING | red | — |
| box_widthopt | INT | 21–10 | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |
| visualization | IMAGE | — |