Midori no Kaze (Green Wind)

Hi again and welcome to Midori no Kaze (also translated as Green Wind).

"Midori no Kaze" is part of my personal color series where I aim to create an specific style but oriented mostly for certain color and component aesthetics so it can be used for characters, objects, landscapes, etc.

"Midori no Kaze" aims to create Japanese/anime generations but in a green/white duotone (sometimes blueish too due to the nature of training daa) combination. This LORA is extremely good for anime line renders, dynamic compositions, stylish standoffs, anime characters reimaginations, cybernetic components on characters, certain components of the Japanese culture, etc.

Main token is mdrnkz3 and class is style so you can use mdrnkz3 style or just mdrnkz3. I recommend the first option.

This lora also has powered sub-tokens as: "white background", "green" (careful with this), "cyber/mecha", "flow/flowing" "wind", (again careful), etc.

Highly recommended to NOT use easynegative embedding due to the training and it's integrity preservations.

Careful too using kl-f8-anime2 VAE cause even looks wonderful due to the nature of the training sometimes could be better without (as delete blueish colors), comparison:

How to use this LORA?
*************************

You can refer to the sample images I upload in the post but:

- For characters mdrnkz3 style Portrait/side portrait of concept... / a mdrnkz3 style concept...

- For landscape and objects a [concept/object/whateveryouwnt] in the style of mdrnkz3

Which are the best values?
*******************************

- Extremely recommended 1 but you can use values from 0.8 onwards. Didn't test enough at higher than 1 but I don't recommend to go further than 1.1 or so.
- CFG Scale can go from 6 downwards. 6 is a default standard value and should be more than OK but you can experiment with 3-6.

Which models are good for this LORA?
********************************************

- Tested on: Cardos Anime, Colorful, Rev Animated, Dreamshaper, Ghostmix, etc. but you will have good outcomes using anime integrated models (almost all, yes). Clip Skip 1 also is allowed, yes but maybe better 2.

This LORA is GOOD for:
******************************

- Character illustrations.
- Mechanical or cybernetic nature/animals.
- White/Green aesthetics towards Anime style (Clip Skip 2 preferred option)
- Dynamic and flowing elements.
- Standoffs
- Characters wearing cybernetic armor/suit/whatever
- Eastern culture landscapes (minimal)
- Anime lineworks
- Anime posters

IMPORTANT!

I wrote an article here and it's like my finetunning diary for the ones interested with releases upcoming, trainings done, exclusive patreon models previews, personal thoughts, etc. So if you want to see what I'm planning or what I'm into, feel free to take a look.

As always thanks to everyone reviewing, giving feedback , commenting and supportin us in Patreon. Keep supporting us so we can keep creating and delivering quality content. Please support the Patreon since it's making this things to be possible around here.

David