Sign In

Esthetic Lina Inverse

604
3.0k
14
Verified:
SafeTensor
Type
LoRA
Stats
2,966
Reviews
Published
Feb 10, 2023
Base Model
SD 1.5
Trigger Words
lina inverse
Hash
AutoV2
83A6E6A1FA
default creator card background decoration
404 Image Contest Participant
OneRing

Lina inverse is mostly 90s anime art, little fan art, mostly anime footage and official ones to keep Lina's style and features.

P.P.S. new model ogipote (2.5D) and ogipote (2.5D)_v2.0 on Boosty

I made a video on how to achieve the same quality or just repeat my art

Lina Inverse Remake on Patreon

CFG Scale - tested on previous version and 4-6 works best

Use almost any weight, from 0.4 to 1 depending on your preferences, the lower the weight, the more variety in postures, clothes and facial expressions, as well as better hands, but more weight also works + Lina's features are much better. Below 0.4 renders mostly hair and small aspects of clothing, but not the outlines of the eyes and face.

Can be used not only to display Lina, but in general, if you do not use Lina-related hints, such as: armor, cape, earrings, gloves, headband, jewelry, lina inverse, boots, then you can achieve the aesthetics of the 90s and the style of E&G Anime Studio (I will add a couple of examples, but if you do not use a weight of 0.9-1, then the results will be much more interesting)

just use hires fix as much as your video card is capable of, so as not to give an error, with 8gb of video memory it is 512x768 with hires fix 2, upscaler Latent (nearest-exact) or just Latent (if there are problems with the image), the picture will be much more detailed.

Use promt 1990s \(style\) to enhance the 90s aesthetic

saving negative prompts to reduce the number of signatures: artist name, signature, watermark,

Trained on AbyssOrangeMix2_hard, you can try to generate on grapefruit, but as for me, AbyssOrangeMix2_hard is a little better.

Combined with my other LORA

How to use LoRA's in auto1111:

  • Update webui (use git pull like here or redownload it)

  • Copy the file to stable-diffusion-webui/models/lora

  • Select your LoRA like in this video

  • Make sure to change the weight (by default it's :1 which is usually too high)

    *Information taken from Lykon

    Don't forget to leave your work or wishes, it will greatly help me improve!

    If you like my work, click on the heart above, I will be pleased :3