Guofeng 3 GuoFeng 3

I solemnly declare: This model is prohibited from being used to train style models based on portraits of celebrities or public figures, as this will bring controversy and have a detrimental negative impact on the AI community.

This model specifies: No real human materials are included in the training materials.

Recommended realistic version of national style:https://civitai.com/models/77650/guofengrealmix

Some Q&A

TIP: GuoFeng 2 or GuoFeng 3 and sub-versions under the main version are independent existences, different versions of the painting style, keywords, and corresponding content will have corresponding differences. It is recommended to download according to the effect images or the independent introduction of the model.

TIP2: The difference between GuoFeng 2 and 3: I classify the main versions differently based on training quality (such as material resolution, the amount of memory consumed by training settings, and sometimes also related to the amount of material).

===========

Update History

2023.6.29 added v3.4 version

2023.4.05 added v3.3 version

2023.3.02 added v3.2_light version

2023.2.27 added safetensors format

-

2023.6.29 added v3.4 version

2023.4.05 added v3.3 version

2023.3.02 added v3.2_Light version

2023.2.27 added safetensors format

===========

Model Introduction

Welcome to use GuoFeng3 model - This is a Chinese gorgeous ancient style model, also can be said as an ancient style game character model, which has a 2.5D texture.

GuoFeng3: Original model

GuoFeng3.1: Fine-tune repair on GuoFeng* *ortrait

GuoFeng3.2: If you are unsure whether to choose GuoFeng3 or GuoFeng2, you can directly use this version

GuoFeng3.2_light: fused with Lora trained on Noise Offset through GuoFeng3.2 to create more beautiful lighting effects (Lora: epi_noiseoffset/Theovercomer8's Contrast Fix)

GuoFeng3.3: This version is a significant update and improvement based on 3.2, it can adapt to full body, even if your tags are not very good, the model will automatically modify the image, but this may lead to similar faces being produced by the model.

GuoFeng3.4: This version has been retrained for full-body images, with significant differences in content from the previous versions. The overall painting style has been adjusted, reducing overfitting and allowing for more adjustments using Lora for images and content.

Other personal models:

Similar model style Lora version: https://civitai.com/models/11352/guofeng3lora

Western fantasy model: https://civitai.com/models/10818/westmagic

===========

GuoFeng3 Optional Lora - GuoFeng3DLC(Lora)

1. Dream (https://civitai.com/models/44310/dream-based-on-guofeng3)

2. WuMo (https://civitai.com/models/47728/wumo)

===========

Recommended settings

Sampling steps: 30 or 50

Sampler: DPM++ SDE Karras

VAE: vae-ft-mse-840000

stabilityai/sd-vae-ft-mse-original at main (huggingface.co)

Keywords:

best quality, masterpiece, highres, 1girl, china dress, Beautiful face

Negative words:

NSFW, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, bad feet

Better negative words:

(((simple background))), monochrome, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, lowres, bad anatomy, bad hands, text, error, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, ugly, pregnant, vore, duplicate, morbid, mutilated, transsexual, hermaphrodite, long neck, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, bad anatomy, bad proportions, malformed limbs, extra limbs, cloned face, disfigured, gross proportions, (((missing arms))), (((missing legs))), (((extra arms))), (((extra legs))), pubic hair, plump, bad legs, error legs, username, blurry, bad feet

If you want more elements, you can add the following keywords:

Beautiful face, hair ornament, solo, looking at the viewer, smile, closed mouth, lips, china dress, dress, hair ornament, necklace, jewelry, long hair, earrings, Chinese clothes, architecture, east Asian architecture, building, outdoors, rooftop, city, cityscape

Suggested keywords for this version:

best quality, masterpiece, highres, cg, //Your tag location, lighting, candid, Photograph, high resolution, 4k, 8k, Bokeh,

This version has been retrained for full-body images, with significant differences in content from the previous versions. The overall painting style has been adjusted, reducing overfitting and allowing for more adjustments using Lora for images and content.