This LoRA was mainly tested with the model "grapefruitV3.2" and "kl-f8-anime2" VAE, and it is intended that you run it alongside your main model using "kohya-ss sd-webui-additional-networks extension" for auto1111. (You can experiment with other models, I'm sure.) I recommend 0.6-0.8 weight. The example images are without inpainting for generation references reasons. This is also my first LoRA.
Hope you enjoy!
Description
FAQ
Comments (9)
Cant get this to work with anythingv4.5, it doesnt generate anything even remotely similar to the pics, not sure what im doing wrong since every other lora works for me Sadge
Hmm... I don't know what could be going wrong for you. I tested it with anything v4.5 and it gens perfectly fine. I also added that gen at the end of the example images. You could try to copy the generation data of said gen and see if it works?
@blondebro it gives me this error: Error running process: /content/stable-diffusion-webui/extensions/sd-webui-additional-networks/scripts/additional_networks.py
Traceback (most recent call last):
File "/content/stable-diffusion-webui/modules/scripts.py", line 338, in process
script.process(p, *script_args)
File "/content/stable-diffusion-webui/extensions/sd-webui-additional-networks/scripts/additional_networks.py", line 177, in process
network, info = lora_compvis.create_network_and_apply_compvis(du_state_dict, weight, text_encoder, unet)
File "/content/stable-diffusion-webui/extensions/sd-webui-additional-networks/scripts/lora_compvis.py", line 73, in create_network_and_apply_compvis
network_dim = min([s for s in size if s > 1])
ValueError: min() arg is an empty sequence
@blondebro basically it just generates without the lora for some reason, im using a google colab amything4.5 setup and every other lora worked for me perfectly
Can you make Xayah? with prompted hood and hoodless and clothing variety heart
Probably not, seeing that there is already a pretty decent LoRA out there of her already. 😥
wish it was easier to collaborate on civit. I shared the dataset on the Xayah page and I'd be open to having shared authorship of it so that others could improve on it.
it'd be cool if you could train on LoRAs on civitai, since it's really not that complicated to train a LoRA. just need the hardware.
@lolcharacters if only i could help, but not enough vram. i'm ripping my blackhole on my back to make some images, prompting everything and praying to the matrix guð
@lolcharacters That would be intriguing for sure!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.