Qwen-Image-Edit-2509_clear

• Guide (External site): English | Japanese
• Civitai: Base Model | VAE

A fine-tuned version of Qwen-Image-Edit-2509 model, designed to produce clearer and more vibrant outputs.

Compared to Qwen Image Edit 2509 (Original)

Qwen-Image-Edit-2509_clear | Qwen-Image-Edit-2509 (Original)

Compared to the original, this model generates illustrations with higher contrast, richer colors, and more refined details.

Compared to Google Nano Banana

Restore a black-and-white photo (Original: Audrey Hepburn's public domain photo)

Qwen-Image-Edit-2509_clear | nano-banana

Anime Illustration

Google nano-banana excels in prompt understanding and illustration variation, but Qwen-Image-Edit-2509_clear is superior when it comes to photorealistic texture rendering.

qwen_image_clear_vae

An improved version of qwen_image_vae.

Compared to qwen-image-vae (Original)

qwen_image_clear_vae | qwen_image_vae (Original)

Compared to the original, it produces slightly higher saturation and healthier, more vibrant skin tones.

Please use it according to your preference.

Workflow

Text to Image

This PNG image contains a ComfyUI workflow.

Image to Image

Recommended Settings

General

• CFG scale: 2.5
• Sampler:
  • euler (Speed)
  • heunpp2 (Quality)
• Scheduler: beta
• Steps: 14 (12-24)
• ModelSamplingAuraFlow: shift 4 (3-6)

Upscaler

your choice, examples:

• 4x-SwinIR-L_GAN
• 4x_NMKD-Superscale-SP_178000_G

LoRAs

Qwen-Image-Edit-2509_clear model is only a fine-tuned version of Qwen-Image-Edit-2509, we believe that LoRAs designed for Qwen-Image-Edit-2509 should still be effective.

---

Tips (External site)

• ComfyUI VRAM management: English | Japanese
• Compare BF16, FP8, GGUF in Flux.1: English | Japanese
• Steps and Scheduler: English | Japanese
• Image Editing AI models: English | Japanese

---

License

• Apache-2.0