AI-KSK ComfyUI workflow: Qwen Image 2.1 Image Enhancement.
This workflow is the Qwen Image 2.1 image enhancement (xi-tu) pipeline from the 15-pair test series: it takes any illustration or real photo and redraws it as a cleaner, sharper, higher-detail image while preserving the original composition, lighting, and atmosphere. It runs on the Qwen-Image 2.1 BF16 base model with the Detail LoRA at 0.5 strength, so the source style and framing stay intact while detail and realism go up.
Main features:
- Qwen-Image 2.1 BF16 base model with Qwen3-VL 8B FP8 text encoder
- elusarcas-qwen2-1-detailer-v1 LoRA at 0.5 strength on the model path
- 25-step euler schedule, 1024x1024 output
- Built-in enhancement prompt that preserves composition, lighting, and style
- Single SaveImage output, no bypassed branches
Suggested workflow:
Feed any source image you want enhanced - anime render, real photo, or generated art. Run it as a second pass after your main generation to lift detail without re-rolling the composition. If the result drifts from the original style, lower the LoRA strength; if it looks too close to the input, raise it toward 0.7.
RunningHub workflow (try online, no install needed):
https://www.runninghub.ai/zh-cn/post/2103123291263606786?inviteCode=rh-v1111
AI-KSK RunningHub workflow hub:
https://www.runninghub.ai/user-center/1892566146468511746?inviteCode=rh-v1111
Bilibili tutorial:
https://www.bilibili.com/video/BV1SxaT6nEvk/
Viggle/Qwen-Image-2.1-viggle-turbo:
https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo
RunningHub registration and AI tools:
https://www.runninghub.ai/?inviteCode=rh-v1111
https://www.runninghub.ai/ai-generator/2046514150500524034?inviteCode=rh-v1111
https://www.runninghub.ai/vip-rights/3?inviteCode=rh-v1111
https://rhtv.runninghub.ai/projects?inviteCode=rh-v1111
Stable global access:
https://lelians.com/#/register?code=mFIJ5tZn
TokenRhythm AI model credits:
https://tokenrhythm.studio/i/rf_tr_kM6wekDquDDZQD4PUNmgiuHA
Youyun Cloud GPU:
https://passport.compshare.cn/register?referral_code=A7fTQUYY9lCDFJwA6MMvoq
https://www.notion.so/2b72459a3ac6808f8282f2c687ed8f52
RunPod Cloud GPU:
https://runpod.io/?ref=sxk5n5m7
Support AI-KSK:
https://ko-fi.com/aiksk
Description
AI-KSK ComfyUI workflow: Qwen Image 2.1 Image Enhancement.
This workflow is the Qwen Image 2.1 image enhancement (xi-tu) pipeline from the 15-pair test series: it takes any illustration or real photo and redraws it as a cleaner, sharper, higher-detail image while preserving the original composition, lighting, and atmosphere. It runs on the Qwen-Image 2.1 BF16 base model with the Detail LoRA at 0.5 strength, so the source style and framing stay intact while detail and realism go up.
Main features:
- Qwen-Image 2.1 BF16 base model with Qwen3-VL 8B FP8 text encoder
- elusarcas-qwen2-1-detailer-v1 LoRA at 0.5 strength on the model path
- 25-step euler schedule, 1024x1024 output
- Built-in enhancement prompt that preserves composition, lighting, and style
- Single SaveImage output, no bypassed branches
Suggested workflow:
Feed any source image you want enhanced - anime render, real photo, or generated art. Run it as a second pass after your main generation to lift detail without re-rolling the composition. If the result drifts from the original style, lower the LoRA strength; if it looks too close to the input, raise it toward 0.7.
RunningHub workflow (try online, no install needed):
https://www.runninghub.ai/zh-cn/post/2103123291263606786?inviteCode=rh-v1111
AI-KSK RunningHub workflow hub:
https://www.runninghub.ai/user-center/1892566146468511746?inviteCode=rh-v1111
Bilibili tutorial:
https://www.bilibili.com/video/BV1SxaT6nEvk/
Viggle/Qwen-Image-2.1-viggle-turbo:
https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo
RunningHub registration and AI tools:
https://www.runninghub.ai/?inviteCode=rh-v1111
https://www.runninghub.ai/ai-generator/2046514150500524034?inviteCode=rh-v1111
https://www.runninghub.ai/vip-rights/3?inviteCode=rh-v1111
https://rhtv.runninghub.ai/projects?inviteCode=rh-v1111
Stable global access:
https://lelians.com/#/register?code=mFIJ5tZn
TokenRhythm AI model credits:
https://tokenrhythm.studio/i/rf_tr_kM6wekDquDDZQD4PUNmgiuHA
Youyun Cloud GPU:
https://passport.compshare.cn/register?referral_code=A7fTQUYY9lCDFJwA6MMvoq
https://www.notion.so/2b72459a3ac6808f8282f2c687ed8f52
RunPod Cloud GPU:
https://runpod.io/?ref=sxk5n5m7
Support AI-KSK:
https://ko-fi.com/aiksk