Skip to content
WorkflowListedExternal
Open source

[FLUX] Florence2 img2img basic workflow (for beginner) — v1.0

This workflow demonstrates a fundamental setup for img2img processing using Flux and Florence2.

Sourcecivitai.com/models/1035986
TaskImage to imageFamilyFLUXEstimated VRAMNot yet characterizedBgrade

Runtime profile

Access
External workflow
Status
Listed
Source
Civitai
Updated
Oct 2, 2026

Source description

This workflow demonstrates a fundamental setup for img2img processing using Flux and Florence2.

While it doesn't include any advanced features, it serves as a crucial example showing how Florence2 extracts text from images. Understanding this basic structure allows users to build more complex and creative workflows by integrating various nodes.

It's a valuable learning tool for beginners.

If you don't have the Florence2 node, you can easily install it through ComfyUI Manager using the 'Install Missing Custom Nodes' feature.

=======================================================

이 워크플로우는 Flux와 Florence2를 사용하여 img2img 작업을 수행하는 기본적인 설정을 보여줍니다.

특별한 기능은 없지만, Florence2가 이미지에서 텍스트를 추출하는 과정을 명확히 보여주는 중요한 예시입니다. 이 기본 구조를 이해하면 다양한 노드와 연계하여 더 복잡하고 창의적인 워크플로우를 구축할 수 있습니다. 초보자에게 유용한 학습 도구가 될 것입니다.

Florence2 노드가 없다면 ComfyUI Manager에서 'Install Missing Custom Nodes' 기능을 사용하여 쉽게 설치할 수 있습니다

A1 [contact & follow & like & Buzz]

Mail : a01demort@gmail.com

CIVITAI : https://civitai.com/user/a01demort

Youtube : https://www.youtube.com/@A01demort

Instagram : https://www.instagram.com/a01demort/

Naver Cafe : https://cafe.naver.com/sdfkorea

sdhotline : https://sdhotline.com/

What you need

Model files
21.2 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

11

  • clip_l.safetensors

    Possible

    Text encoder · Unknown

  • DualCLIPLoader

    Not resolved

    Text encoder · Unknown

  • DualCLIPLoaderGGUF

    Not resolved

    Text encoder · Unknown

  • FLUX_VAE_diffusion_pytorch_model.safetensors

    Possible

    VAE · Unknown

  • flux1-dev-fp8.safetensors

    Possible

    Checkpoint · Unknown

  • flux1-dev-Q8_0.gguf

    Found

    Checkpoint · 11.8 GB · GGUF · fp16 · Unknown

  • microsoft/Florence-2-base-ft

    File unverified

    Vision model · Hugging Face · microsoft/Florence-2-base-ft

  • t5xxl_fp16.safetensors

    Possible

    Text encoder · Unknown

  • VAEDecode

    Possible

    VAE · Unknown

  • VAEEncode

    Not resolved

    VAE · Unknown

  • VAELoader

    Not resolved

    VAE · Unknown

Node packs

17

  • BasicGuider

    Not resolved

    Node pack · Unknown

  • BasicScheduler

    Possible

    Node pack · Unknown

  • CLIPTextEncode

    Possible

    Node pack · Unknown

  • DownloadAndLoadFlorence2Model

    Possible

    Node pack · Unknown

  • Florence2Run

    Possible

    Node pack · Unknown

  • GetImageSizeAndCount

    Possible

    Node pack · Unknown

  • Image Comparer (rgthree)

    Not resolved

    Node pack · Unknown

  • KSamplerSelect

    Possible

    Node pack · Unknown

  • LoadImage

    Possible

    Node pack · Unknown

  • PreviewImage

    Possible

    Node pack · Unknown

  • RandomNoise

    Not resolved

    Node pack · Unknown

  • SamplerCustomAdvanced

    Not resolved

    Node pack · Unknown

  • SaveImage

    Possible

    Node pack · Unknown

  • ShowText|pysssss

    Found

    Node pack · Unknown

  • Text Find and Replace

    Found

    Node pack · Unknown

  • UNETLoader

    Possible

    Node pack · Unknown

  • UnetLoaderGGUF

    Possible

    Node pack · Unknown

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • This beginner workflow is for basic img2img processing with Flux and Florence2: it takes an image through LoadImage, can extract text from that image with Florence2, and includes preview, image-saving, and text-display steps.

Flow

  • The listed processing path includes LoadImage, VAEEncode, SamplerCustomAdvanced, Florence2Run, VAEDecode, PreviewImage, and SaveImage.

  • It is presented as a basic example for learning Florence2 text extraction from images, without advanced features.

Inputs and outputs

  • Input: an image loaded through LoadImage.

  • Named output steps are PreviewImage, SaveImage, and ShowText|pysssss.

Setup

  • If Florence2 is missing, use ComfyUI Manager’s “Install Missing Custom Nodes” feature to add it.

    Suggestion · not verified

Dependencies

  • Listed model files are clip_l.safetensors, FLUX_VAE_diffusion_pytorch_model.safetensors, flux1-dev-fp8.safetensors, flux1-dev-Q8_0.gguf, microsoft/Florence-2-base-ft, and t5xxl_fp16.safetensors.

  • The workflow lists BasicGuider, BasicScheduler, CLIPTextEncode, DownloadAndLoadFlorence2Model, DualCLIPLoader, DualCLIPLoaderGGUF, Florence2Run, GetImageSizeAndCount, Image Comparer (rgthree), KSamplerSelect, and LoadImage.

  • It also lists PreviewImage, RandomNoise, SamplerCustomAdvanced, SaveImage, ShowText|pysssss, Text Find and Replace, UNETLoader, UnetLoaderGGUF, VAEDecode, VAEEncode, and VAELoader.

Configuration

  • The listed base model is Flux.1 D, and the listed model names include flux1-dev-fp8.safetensors and flux1-dev-Q8_0.gguf.

Cautions

  • Before using it, verify that Florence2Run, UNETLoader, SaveImage, and Text Find and Replace are available in ComfyUI.

    Suggestion · not verified

  • No license is listed; check applicable usage terms before distributing the workflow or its outputs.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimary
civitai.com/models/1035986

Verified Jul 20

Source excerpts

1 excerpt
NoteJul 20civitai.com/models/1035986

Source context: 559 downloads · Type Workflows · Base model Flux.1 D