Refine and detail for SD1.5, SDXL, and PONY.
Best model blends [base] + [refiner]:
Pony realistic (semi-realistic) + realistic
Cartoon/Anime + Pony realistic (semi-realistic)
Realistic + Photorealistic/dedicated extreme detail
You can blend model architecture, i.e., SD1.5 + PONY; I suggest you turn the denoise up in the highres fix to smooth out artifacts this tends to produce. If you do this, the VAE will not match, and the preview image will look burned. You may need to add a new load VAE node and pipe it into the corresponding 'to basic pipe' node.
Posting images is appreciated. Reach out if you need help troubleshooting.
Description
Added detailers and spread out nodes to make it easier to follow. Added quality of life nodes.
FAQ
Comments (7)
Excellent work. It improves the hires quality significantly. Better than A1111 text2img hires now.
Thanks for the excellent and creative Pony workflows. Just what I was looking for.
Could you tell me which lora this is?
'XL\Add details.safetensors'
to bad it dont work with the new comfyui version
i may be hit by noobiness but I tried this workflow. After correcting the different models etc. and running it.. it finishes in 4 seconds and tell me it created "stringfunction|pyssss"
Nothing else happens.
What am I failing to understand?
Your workflow is amazing for beginners, just plug and play. Photorealistic results without the hassle, thanks!
Vae is literally broken in all of the worfklow are you guys insane?
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.