High five

Statement: To prevent malicious use on other platforms, the model will be available for irregular download, and the model will continue to run.

For communication and sharing, welcome to join the AI Mutual Aid Group, search QQ group number:482570609

→→→→→→Get more high-quality models←←←←←←

This Lora will randomly generate a pair of people clapping in the image,

from lovers to best friends, to beauty and the beast... see what you will randomly get.

Sometimes it will also create some warm moments, very suitable as a New Year greeting image.

This artist also wishes everyone a Merry Christmas in advance with this work! Happy New Year!

As a New Year's gift, this Lora has no restrictions on downloads, hope everyone has fun playing with it.

Instructions:

Sweetness weight at 0.6, this Lora has good generalization ability, supports most large drawing models, realistic models have not been tested yet.

As it is a multi-character scene, it is recommended to have a resolution above 768*768. For high image quality, upscaling and ADetailer are needed.

The clapping characters are randomly generated based on the content covered by the large model you choose.

You can modify the trigger words to change gender, for example, change 2characters to 2boys, 2girls. For specifying characters, you need to use LatentCouple with the corresponding character Lora.

Recommended Parameters:

parameters

highfive, 2 characters, happy new year,
dynamic pose, city, at night, Chinese elements, anime,
<lora:HighFive:0.6>,
Steps: 30, Sampler: DPM++ 2M Karras, CFG scale: 6, Seed: 1220838831, Size: 800x800, Model hash: 4199bcdd14, Model: ReV Animated_v1.22, Denoising strength: 0.39, Clip skip: 2, ADetailer model: face_yolov8s.pt, ADetailer prompt: detailed beautiful face, ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 23.11.1, Hires upscale: 1.5, Hires steps: 9, Hires upscaler: R-ESRGAN 4x+ Anime6B, Lora hashes: "HighFive: 57c0db20062b", Version: v1.6.0-2-g4afaaf8a

This is the artist's first attempt at training dynamic Lora. Please provide feedback for improvement.

Respect copyright, respect individual labor results, please credit the author when using this model elsewhere. This model is for learning and communication purposes only, do not use for any commercial activities.