I DO NOT authorize PixAI & Yodayo (and other users) to transfer this model from Civitai.
PixAI users who wish to use this model can do it by using the uploaded one on my PixAI profile from the link below : https://pixai.art/@vxp/artworks/models
Attention: this is a vPred model.
Use environments compatible with this technology, such as ComfyUI, Forge, etc.
The model has been reworked to produce Anime-style images, but at the same time, by using prompts like "realistic, photorealistic" it is capable of generating realistic images, thereby expanding the range of creativity.
The following settings are suggested:
SAMPLER: Euler / Euler A / DPM++ and others
SCHEDULER: normal
STEPS: 20 (16 ~ 25, as you prefer)
CFG: 5 (4.0 ~ 6.0)
Recommended quality prompts:
masterpiece, best quality, amazing quality, absurdres, extremely aesthetic,newest
Recommended negative prompts:
bad quality,worst quality,worst detail,bad, low quality,low details, poorly drawn, bad anatomy, extra fingers, bad hands, bad finger, bad fingers, split fingers, missing finger, split finger, split fingers, extra digits, fused arms, fused hands,split limbs
This model has the following characteristics:
– High prompt-following capability (inherited from illustriousXL v2.0), full compatibility with classic TAG-style prompting, and support for natural-language style descriptions.
– Strong flexibility when generating images with extreme aspect ratios (for example 5:1 to 1:5).
– Support for a wide range of graphical and artistic styles.
– Excellent visual quality, suitable for both anime and semi-realistic images, with accurate rendering of small decorative details.
– Most importantly, being a vPred model enables the generation of images with high contrast. with vPred
All image uploaded by me was generated WITHOUT using any kind of enhancing, it’s mean no HiRes, no upscale, no detailer or other stuff, just a raw output.