CivArchive
    Preview 1
    Preview 2
    Preview 3
    Preview 4
    Preview 5
    Preview 6
    Preview 7
    Preview 8
    Preview 9
    Download and instruction here: https://civitai.com/models/317902?modelVersionId=422446 This model perfectly supports lora trained with ponyv6 as the base model, and the Lora of ani3 and sdxl1.0 can also be adapted to some extent. pony是神,兼容性满分。本模型支持ani,pony的lora 必备前置效果词和ponydiffusion一样 positive:(score_9,score_8_up,score_7_up,score_6_up,score_5_up,score_4_up) OR (score_9,score_8_up,score_7_up)

    Description

    Rights/Usage/Uploads all at bottom (This is not my model) This model belongs to Civitai User: Tonade Their profile: https://civitai.com/user/Tonade The latest version of the model: https://civitai.com/models/317902?modelVersionId=805539 Creators Notes: This version has added more scene materials, which can generate more beautiful scene images and better scene composition compared to previous versions. It has also optimized the finger problem that existed in v5.5 and the toe problem of the foot under some prompt words. V6 usage tip: The word "anime" is used to train the art style of nai3. If European and American art styles appear during use, this word can be used to stabilize the nai3 art style. This is a method I tried to reduce overtrain. - The model already has included vae, there is no need to add additional vae - The best generate strategy is to use high-fix at a moderate resolution, rather than directly using high-resolution direct output - A link to their discord: https://discord.com/invite/CXKvxTSfkm

    FAQ