use 0 images for finetune XLbase,should fit any XL model(even anime)
add detailed eyes、detailed face and detailed skin
including neg part and fixed bad hands
Less effect over 75 tokens
Description
add more eyes details
FAQ
Comments (17)
When trying to merge your Lora I get:
no module found for LoRA weight: lora_unet_up_blocks_1_attentions_2_transformer_blocks_1_attn2_to_q.lora_down.weight
This is true for both V3 and V2 but not V1. Do you have any idea what it can be? :)
i think it is because of diffuser version...
kohya merge script dont full support diffuser training lora
@bdsqlsz I see! Any tips on merging it outside of Kohya? If not, thanks anyways for the answer, I was banging my head against the wall :P <3
@olivetty you can just use supermerge.
@bdsqlsz Tried it but didn't work. Got TypeError: unsupported operand type(s) for +=: 'Tensor' and 'tuple'
Latest 1.6.0 A1111 and latest commit for the Supermerger extension.
@olivetty use 1.5.0 will be OK,lastest A1111 cant support supermerge
This LoRA brings great effect to eyes but it seems anti-nsfw with XL checkpoints. Any ideas?
i dont use any sfw datasets...
it is strange
The only time I've ever experienced the "anti-NSFW" effect with part-specific LoRAs is when the strength is turned up past 100%. Sometimes turning down the strength ratio fixes this. I usually initiate my detail LoRAs during the upscaling phase to improve workflow efficiency. This has the added benefit of dramatically reducing probability of the "anti-NSFW" effect.
Could you just change models at the final step then run ADetailer?
Can you make this available for the generation services? It seems like at times I can find it, but most of the time it doesn't seem available.
This lora is my favorite for eyes detailer, and hey, it also add subtile detail in the image other than eyes too. I love it
Thank you for comment!
@bissonfrederic199 is right, I use this lora all the time for general improvements, even when my prompts wouldn't expect to produce an image with eyes. Great lora, thank you for sharing!
So, what prompt do I have to use? It didn't work when I used the one shown in the picture
Details
Files
DetailedEyes_V3.safetensors
Mirrors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
1-EyesDetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
2.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
28-detailedeyes.safetensors
DetailedEyes_V3.safetensors
DetailedEyesXL.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
detailedEyes_v3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
4ee2df15-54d5-412e-b265-da21c7bcf192.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
DetailedEyes_V3.safetensors
Available On (4 platforms)
Same model published on other platforms. May have additional downloads or version variants.



