# kijai/ComfyUI-Florence2

Inference Microsoft Florence2 VLM

Repository: https://github.com/kijai/ComfyUI-Florence2
Canonical: https://ross.abutalabs.com/products/comfyui-florence2
Language: Python
License: MIT
License Family: permissive
Last push: 2026-05-06T09:00:47+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 81, release rhythm 35, longevity 57
- inputs: {"age_days": 805, "days_push": 119, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1742, forks 139 (observed 2026-08-28T04:05:30.394470+00:00)

## What it is
A ComfyUI custom node plugin that runs Microsoft's Florence-2 vision-language model for image captioning, object detection, segmentation, and document visual question answering. It supports automatic model downloads and various community fine-tunes such as PromptGen and SD3/Flux captioners.

## Use cases
- generate captions for images inside comfyui workflows
- run object detection and segmentation with florence-2
- ask questions about scanned receipts and documents
- generate stable diffusion prompts from images
- batch caption images for dataset preparation
- extract text information from forms and letters

## When to choose
- you already use ComfyUI and want vision-language inference as nodes
- you need captioning, detection, or DocVQA without writing code
- you want to try community Florence-2 fine-tunes like PromptGen

## When to avoid
- you need a standalone CLI or API server rather than ComfyUI nodes
- you don't use ComfyUI and only want raw Florence-2 inference
- you need training or fine-tuning of Florence-2 itself

## Facets
- artifact type: plugin
- maturity: active
- function: machine-learning, computer-vision, nlp, ocr, image-processing
- domain: computer-vision, artificial-intelligence, image-processing
- platform: python, cross-platform
- tags: comfyui, custom-node, florence2, vlm, vision-language-model, captioning, object-detection, docvqa, stable-diffusion-workflow, natural-language-processing, desktop

## Member repositories
- kijai/ComfyUI-Florence2 (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:30.394470+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:29:55.651175+00:00, confidence not recorded.
  - readme: https://github.com/kijai/ComfyUI-Florence2 (fetched 2026-08-28T04:05:30.394470+00:00, sha f315dd0d5ea5)
- Data as of 2026-08-30T08:39:29.467469+00:00.