Free ComfyUI Workflow: Promptless Style Transfer with Florence-2

Tired of endlessly tweaking prompts to capture the perfect artistic style? This revolutionary ComfyUI workflow lets you transfer any style onto any image—without writing a single word. By harnessing the power of ByteDance's new Florence-2 model, this setup automates the entire creative process, letting you focus on the vision, not the vocabulary.

Forget guesswork. Just provide a source image and a style reference, and the workflow intelligently blends them into a stunning, new creation. It's the most intuitive and powerful style transfer tool for ComfyUI.

The Magic Behind the Promptless Method: Florence-2

So, how does it work without a prompt? The secret is Florence-2, a state-of-the-art vision-language model. Instead of you describing the source image, Florence-2 "looks" at it and automatically generates a detailed, accurate text description.

This auto-generated caption becomes the prompt, perfectly capturing the content of your image. The workflow then combines this description with the artistic style of your chosen reference image, giving you a flawless style transfer every time.

Key Features of This Workflow

• 🤖 Zero Prompting Required: Florence-2 handles the description, so you can stop guessing prompts and focus purely on the visual style.
• ⚡ Rapid Iteration: Not the right vibe? Just pick a new style reference and re-run the workflow in seconds. It's the fastest way to explore different creative directions.
• 🎨 Infinite Creative Range: Transfer any aesthetic you can find—from painterly illustrations and photorealism to bold graphic design and cinematic looks.
• 🖼️ Built-in Style Browser: Use the integrated Civitai Gallery Browser to find and select your style reference images without ever leaving ComfyUI.
• ✨ Consistent Results: Apply the same style reference across a series of images to maintain a cohesive and professional visual identity.

How to Use the Promptless Style Transfer

This workflow is designed for maximum speed and simplicity.

1. Drop in Your Source Image: Load any image you want to transform—portraits, landscapes, products, anything works.
2. Pick Your Style Reference: Use the Civitai Gallery Browser node to find an image with the aesthetic you love. The workflow will use this as the style guide.
3. Click "Queue Prompt": That's it! The workflow automatically generates a description with Florence-2, combines it with your chosen style, and renders a new masterpiece.

Who is This Workflow For?

• Designers & Marketers: Quickly build moodboards, explore brand aesthetics, or generate social media visuals from everyday photos.
• Concept Artists: Rapidly develop the look and feel for a project without getting bogged down in prompt engineering.
• AI Artists & Hobbyists: Create stunning, cohesive image series with a single, unified art direction.
• Anyone Tired of Writing Prompts: If you want to focus on creativity and visuals, this workflow is for you.

Frequently Asked Questions (FAQ)

What is Florence-2? Florence-2 is a powerful vision-language model from ByteDance. It's designed to understand the content of images and perform tasks like captioning, object detection, and in this case, generating a text prompt from an image.

What custom nodes are required? You will need the custom nodes for the Florence-2 model and the Civitai Gallery Browser. These can be installed through the ComfyUI Manager.

Can I edit the auto-generated prompt? Yes! The prompt generated by Florence-2 appears in a standard text box. You are free to edit or add to it before running the queue if you want to fine-tune the output.

Unlock Effortless Creativity

Stop wrestling with words and start creating beautiful images. This workflow removes the biggest barrier in AI art generation, putting infinite stylistic power right at your fingertips.

[Download the Promptless Style Transfer Workflow Now]