QwenEdit Consistence Lora

Introduction

Meet the QwenEdit Consistence Lora, a specialized model designed to enhance your image editing experience with QwenEditPlus and QwenImage Edit. This LoRA focuses on maintaining consistency across your edits, ensuring smoother results. It's particularly useful for workflows that prioritize consistent image generation and editing.

This model helps resolve common pixel shift issues and preserves the QwenVL image encoding capabilities, making your creative process more reliable.

How to Use

• Trigger word: No trigger word is required to activate this LoRA.
• Recommended weight/scale:
  • For working with multiple images, a strength between 0.5 and 1.0 is recommended.
  • For single image edits, try a strength between 0 and 0.5.
  • A specific suggestion of 0.45 to 0.5 is also mentioned, which falls within the single image range.
• Best settings & dependencies:
  • This LoRA is trained for and best used with QwenEditPlus (also known as QwenImage Edit).
  • It requires the Comfyui-QwenEditUtilsV2 custom node.
  • It's designed to be used with the QwenEdit Consistence Edit workflow (versions v4.0 or v2.0 are mentioned).

FAQ

Q: What's the trigger word?

A: No trigger word is required for the QwenEdit Consistence Lora.

Q: What weight should I use?

A: For multiple images, try a strength between 0.5 and 1.0. If you're working with a single image, a strength from 0 to 0.5 is suggested. A specific recommendation of 0.45 to 0.5 is also provided.

Q: What base model is this for?

A: This LoRA is specifically designed for the Qwen base model.

Technical Details

• Base Model: Qwen
• Source: Civitai
• Custom Node Requirement: Comfyui-QwenEditUtilsV2
• Recommended Workflow: QwenEdit Consistence Edit workflow v4.0 (also compatible with v2.0)
• Experience Links: You can find more examples and usage insights via the experience links provided on the original Civitai page.

Gallery