(make sure to lower the lora's weight if you want him without a cloak- thank you nonexistent for bringing that up in the comments i should've tested on that beforehand xD) The trigger words for this lora; the_knight, mouthless, black eyes, two horns, white head, black body, blue cloakAdditional non-trigger words to help with the prompting; mouthless, featureless arms (if you want him generated without fingers), hollow knight, featureless legs----Be aware that CivitAI will not give you the same results as the showcases provided. CivitAI generates things differently from Stable Diffusion and ComfyUI. Put your effort in and work him to your own style and image!----I'm still surprised there was only just a single lora for this dude. You'd think they'd have just as much as Hornet, but no, there's only one! One that I didn't even know was a Hollow Knight lora...Trained on 22 images. Great and compatible with styles and poses, though it depends on what style and/or pose you're relying on, so it's best to get experimental. Lora's also great at doing shortstack bodies.Not good for multiple characters, though; Chances they'll either have a missing horn, one horn, weird looking horns, a white body, or no eyes at all. Same can happen solo, but less likely, and has a higher chance of happening depending on the sample (ex: Euler and DDIM). Of course, it may work for you and just not for me. Only see for yourself.Other than that, enjoy the lora y'all.

The Knight / Little Ghost (Hollow Knight)

V1. Expected to be retrained if I ever find more compatible images to train with.

tklg_hk_000001.safetensors

v1.0

(make sure to lower the lora's weight if you want him without a cloak- thank you nonexistent for bringing that up in the comments i should've tested on that beforehand xD) The trigger words for this lora; the_knight, mouthless, black eyes, two horns, white head, black body, blue cloakAdditional non-trigger words to help with the prompting; mouthless, featureless arms (if you want him generated without fingers), hollow knight, featureless legs----Be aware that CivitAI will not give you the same results as the showcases provided. CivitAI generates things differently from Stable Diffusion and ComfyUI. Put your effort in and work him to your own style and image!----I'm still surprised there was only just a single lora for this dude. You'd think they'd have just as much as Hornet, but no, there's only one! One that I didn't even know was a Hollow Knight lora...Trained on 22 images. Great and compatible with styles and poses, though it depends on what style and/or pose you're relying on, so it's best to get experimental. Lora's also great at doing shortstack bodies.Not good for multiple characters, though; Chances they'll either have a missing horn, one horn, weird looking horns, a white body, or no eyes at all. Same can happen solo, but less likely, and has a higher chance of happening depending on the sample (ex: Euler and DDIM). Of course, it may work for you and just not for me. Only see for yourself.Other than that, enjoy the lora y'all.

(make sure to lower the lora's weight if you want him without a cloak- thank you nonexistent for bringing that up in the comments i should've tested on that beforehand xD)

The trigger words for this lora; the_knight, mouthless, black eyes, two horns, white head, black body, blue cloak

Additional non-trigger words to help with the prompting; mouthless, featureless arms (if you want him generated without fingers), hollow knight, featureless legs

----

Be aware that CivitAI will not give you the same results as the showcases provided. CivitAI generates things differently from Stable Diffusion and ComfyUI. Put your effort in and work him to your own style and image!

----

I'm still surprised there was only just a single lora for this dude. You'd think they'd have just as much as Hornet, but no, there's only one! One that I didn't even know was a Hollow Knight lora...

Trained on 22 images. Great and compatible with styles and poses, though it depends on what style and/or pose you're relying on, so it's best to get experimental. Lora's also great at doing shortstack bodies.

Not good for multiple characters, though; Chances they'll either have a missing horn, one horn, weird looking horns, a white body, or no eyes at all. Same can happen solo, but less likely, and has a higher chance of happening depending on the sample (ex: Euler and DDIM). Of course, it may work for you and just not for me. Only see for yourself.

Other than that, enjoy the lora y'all.

Files with SHA256: cec366638f1e75213561b11b0aa694bd7a0e4d0884e887959b06d356aeddcf0d