IAMCCS QE Prompt Enhancer: A Visual Prompt Selector for Qwen Image Editing Models

IAMCCS QE Prompt Enhancer is a custom node designed specifically for ComfyUI, offering a visual button grid interface for quick prompt selection. It is specially optimized for multimodal image editing models like Qwen Image Edit 2509, supporting multiple presets, multi-line prompt workflows, and practical features such as search and append mode, greatly simplifying the AI image editing workflow.

When using multimodal image editing models like Qwen Image Edit 2509, writing prompts is often a time-consuming process that requires repeated trials. The traditional text input method has several obvious issues:

Heavy memory burden: Users need to remember a large number of prompt templates and best practices, which is a high threshold for beginners.

Low input efficiency: Each edit requires manual input or copy-pasting of prompts, especially when frequent switching between different styles or effects is needed.

Difficulty ensuring consistency: Different prompt writing styles may lead to fluctuations in output quality, lacking standardized prompt management.

Lack of visual feedback: Pure text input cannot intuitively show effect previews of different prompt options.

IAMCCS QE Prompt Enhancer was created to solve these problems; it visualizes and standardizes the prompt selection process, making the image editing workflow smoother and more efficient.

The most prominent feature of this node is its intuitive visual interface:

• Preview Cards: Each prompt option has a corresponding preview card, allowing users to intuitively see effect illustrations
• Synchronized Dropdown Menu: Card selection is synchronized with the dropdown menu in real time, catering to different operation habits
• Dark Preview Mode: Thumbnails with black backgrounds and white icons are clearer and easier to read in dark themes.

QE Prompt Enhancer has built-in multiple preset categories covering common image editing needs:

Camera Angle Presets Provides multiple prompt options for professional photography angles, helping users quickly specify perspective requirements.

Dataset Generator Presets Presets (No.1-5) designed specifically for multi-line prompt workflows, suitable for scenarios requiring batch generation of training data. This is an important new feature added in version v1.0.3.

Character Profile Injection Supports injecting character profile information into multi-line prompts, facilitating consistent character editing.

When there are many presets, the search function becomes crucial:

• Instant Filtering: Enter keywords to filter options in the grid in real time
• Quick Positioning: No need to scroll to find; directly jump to the target prompt.

Append Mode: Allows combining selected prompts with settings from other nodes, including consistency control, pose adjustment, re-lighting, QwenVL mode, etc.

Keep Selection: Prevents automatic reset of selection after each execution, suitable for scenarios where the same prompt needs to be used multiple times.

For the QwenVL model, this node provides an exclusive text-side prompt generation function to ensure optimal matching of prompts with the model.

For users with limited VRAM, IAMCCS has specially optimized the Qwen LoRA loader. It is recommended to use the updated version from IAMCCS_nodes to avoid VRAM overflow issues when using Qwen models.