ComfyUI Node: Florence2Run

Authored by kijai

Created

Updated

1718 stars

Run Florence2Run without installing ComfyUI

This node already runs on ComfyICU. Open it in the browser, hit run, and you're billed for GPU seconds — not for setup, and not while you're idle.

Category

Florence2

Inputs

image IMAGE
florence2_model FL2MODEL
text_input STRING
task
  • region_caption
  • dense_region_caption
  • region_proposal
  • caption
  • detailed_caption
  • more_detailed_caption
  • caption_to_phrase_grounding
  • referring_expression_segmentation
  • ocr
  • ocr_with_region
  • docvqa
  • prompt_gen_tags
  • prompt_gen_mixed_caption
  • prompt_gen_analyze
  • prompt_gen_mixed_caption_plus
fill_mask BOOLEAN
keep_model_loaded BOOLEAN
max_new_tokens INT
num_beams INT
do_sample BOOLEAN
output_mask_select STRING
seed INT

Outputs

IMAGE

MASK

STRING

JSON

Extension: ComfyUI-Florence2

Nodes to use Florence2 VLM for image vision tasks: object detection, captioning, segmentation and ocr

Authored by kijai

Looking for a different node?

Also provided by 1 other extension

Run Florence2Run without installing ComfyUI

This node already runs on ComfyICU. Open it in the browser, hit run, and you're billed for GPU seconds — not for setup, and not while you're idle.

Learn more