Raena Qwen Image

Overview

Raena-Qwen-Image is a finetuned lora for Qwen-Image, specialized for anime-style generation. The goal of this model is to enhance Qwen-Image’s capability in producing high-quality anime outputs with more sharper detail, richer color, and better aesthetics. This LoRA is designed to improve Qwen-Image specifically for anime use cases, trained on 500 hand-picked high-quality images with natural caption, making the outputs more visually appealing and aligned with the style expectations of anime art.

Usage

  • Trigger words: start with Anime illustration of . . . and the rest with your prompt.
  • Lora Weights: 0.8 to 1.0
  • Works best when paired with Lightning Lora, allowing for faster and more refined generations.
  • Showcase images include metadata. To get the workflow, simply download the image and drag it into ComfyUI.

Support

If you’d like to support this project or my other developments, you can buy me a coffee on Ko-Fi

Downloads last month
109
Inference Providers NEW

Model tree for Raelina/Raena-Qwen-Image

Base model

Qwen/Qwen-Image
Adapter
(90)
this model