Any illegal act

using this model is prohibited, reprinting at will is prohibited, and only shared as an achievement. Violators are responsible for the consequences! (Fans can receive drawing guides, authorizations, and business cooperation on behalf of Lora plus me VX: 158712235

).

Description: This is a special effects LORA for water [splashes], fire, ice, tornado, milk splash, clouds, and light effects. It is highly sensitive to reminders and can generate various corresponding special effects and creative advertising images. It doesn't affect faces or characters, and you can customize any character to match these special effects elements. It can be used with other characters LORA at the same time, and can generate many different effects, bringing unexpected surprises. [Please use your imagination and post more beautiful pictures] We

recommend the big model: chilloutmix. You can also try other mainstream high-quality real human models, and two-dimensional big models. You will be surprised by drawing a card. The higher the weight, the greater the proportion of special effects elements on the screen, so please adjust as needed

.

Examples of light effect group cues: with chilloutmix large model gradient background, blue light, glowing

, light, masterpiece, best

quality,1girl, gogerous

Dress, clean face, white skin,

Izumo color hint: with the large chilloutmix model, weight 0.6-0.8 chilloutmix

, you can change your favorite LORA 1Girl, gogerous dress,

clean face, white skin, cloud, huge cloud, golden cloud, sky, outdoors, masterpiece, best quality,

as long as milk splashes special effect: LORA, weight 0.8, match

chilloutmix large model

masterpiece, best quality, milk, (milk splash: 1.5), huge milk splash, girl and milk: LORA, weight 0.8,

with large chilloutmix model

Masterpiece, Best Quality, Milk, (milk splash: 1.5), HUGE MILK SPLASH, BLUE BACKGROUND, 1GIRL, WEAR RED HANFU, SMILE, STAND IN MILK GIRL

AND FIRE: LORA, weight 0.8, with chilloutmix large model

masterpiece, best quality, black background, fire,1girl, wear red hanfu, smile, stand

in fire,

Recommended negative terms: (worst quality), (low quality: 1.4), deformed, malformed, deformed face, bad teeth, bad hands, bad fingers, bad eyes, long body, blurry, duplicate, cloned, duplicate body parts, disconnected, extra sparked, fused fingers, extra fingers, twisted, distorted, malformed hands, mutated hands and fingers, conjoined, missing joints, bad anatomy, bad subtitles, logo, watermark, text, copyright, signature, lowres, mutated, mutilated, artifacts, gross, ugly, artist_name, signature, artist_logo, watermark,

nsfw

Number of iterative steps: Adjust to 30-50 steps for a more detailed focus. Recommended sampling methods: DPM++ 2M Karras, DPM++ SDE Karras

, Euler a are highly adaptable.

Key points: enable Adetailer face restoration when drawing faces, select: mediapipe_face_full, or face_yolov8n_v2. Do not check facial restoration. If you are drawing with multiple people, it is highly recommended to start adetailer. It is recommended

to generate 4 images at the same time. You can skip clip and select 2 by default.

If you want to enlarge the image, you can check the high-resolution restoration. The recommended magnification algorithm: 4x-ultrasharp, fix the ideal seed value, and redraw amplitude of about 0.3-0.5.

If you want to make small changes, redraw the profile by 0.3, and

there are many other easy-to-use models: Various models on the homepage [Welcome to use, return to the US] You can view other people's generated information in the return area~ Model Use Statement 1. You must not use this model

or its derivative versions (such as

Fusion model version) hosted on a website/app that plans to earn revenue or donations.

If you need to use this model and its pedestrian version for commercial purposes (such as generative services, selling images, using images for publicly published articles or publications, etc.), please contact me via WeChat (158712235) 2. You may not directly sell images generated by this model. If you violate this section, you are personally responsible for all legal consequences, and I am not responsible for it. You must not use this model to intentionally create or share illegal or harmful output or content; please abide by public order and morals and use this model

for positive and positive purposes.

3. The optimized version will be updated in the future, keep an eye out for news~