Sign In

chosen Irises-mix/鸢尾花

Download

1 variant available

fp16 SafeTensor

chosen irises bake vae.safetensors

Half precision, best balance • 4.11 GB

Verified:

Type
Checkpoint Merge
Stats

5,555

269

Published

Jul 25, 2023

Base Model

SD 1.5

Training
Steps: 10
Epochs: 20
Usage Tips
Clip Skip: 2
Hash
AutoV2
C3FF8D00A6
default creator card background decoration
Followers - 7340

7.3K

Likes - 62918

62.9K

18668-3046214903-(Masterpiece_1.2, high quality), (pixiv_1.4), (1girl_1.2), (solo_1.2),_Elaina,1girl, in container, cocktail glass,full body,_.png

The English instructions are below.

大家好,我是chosen鲷,irises大模型更新至anima版本。

这个模型风格比较清新自然,

我在chosen-mix_anima的基础上进行微调训练得到这个模型。

这个模型风格比较清新,明亮,同时具备了chosen-mix的兼容性和colorbox优秀的色彩表现,,不管人物还是风景都非常优秀,在1.5时代非常受欢迎,所以我把他升级到了anima版本。

推荐风格:

风格1(封面图):masterpiece, score 7, highres, absurdres, best quality, explicit, safe, sfw, detail eyes, multicolored_background, multicolored eyes, high-contrast aesthetic,

@ichika (ichika87),@ningen_mame,@yoshida_akihiko,@ningen_mame,@satsuya,@ciloranko,

风格2:masterpiece, best quality, score_7, score_8, sketch, highres, extremely detailed 8k wallpaper, illustration, depth of field, great quality, amazing quality, very aesthetic, absurdres, detail eyes, multicolored eyes, high-contrast aesthetic,

@ask \(askzy\), @ciloranko, @wlop

推荐这两个,附件里的图包还有许多风格大家可以自行尝试,我所有的lora都可以和这个模型兼容。

生成参数:cfg1 step12(使用加速lora的情况下)cfg4 step50(不使用加速lora的情况下),euler+simple,降噪 0.95,尺寸建议直接出1920*1080。

感谢大家的下载使用,如果喜欢的话欢迎关注我,我的bilibili pixiv ID均为chosen鲷,欢迎大家关注我的频道https://space.bilibili.com/3867754?spm_id_from=333.337.0.0 https://www.pixiv.net/users/102907778 谢谢。

The following content is from machine translation

Hello everyone, I'm chosen Dia. I'm excited to announce that the irises model has been updated to the anima version!

This model features a naturally fresh and bright aesthetic. Built upon my previous chosen-mix_anima, it inherits the robust compatibility of the chosen-mix series while embracing the vivid, brilliant color performance of Colorbox—excelling equally in characters and landscapes. It was incredibly well-received in the SD 1.5 era, so I decided to bring it up to the anima architecture.

Recommended Styles:

Style 1 (cover image):

masterpiece, score 7, highres, absurdres, best quality, explicit, safe, sfw, detail eyes, multicolored_background, multicolored eyes, high-contrast aesthetic, @ichika (ichika87), @ningen_mame, @yoshida_akihiko, @ningen_mame, @satsuya, @ciloranko

Style 2:

masterpiece, best quality, score_7, score_8, sketch, highres, extremely detailed 8k wallpaper, illustration, depth of field, great quality, amazing quality, very aesthetic, absurdres, detail eyes, multicolored eyes, high-contrast aesthetic, @ask \(askzy\), @ciloranko, @wlop

I've included a variety of other styles in the attached image pack—feel free to experiment! All my LoRAs are fully compatible with this model.

Generation Settings:

CFG/Steps: CFG 1, Step 12 (with accelerator LoRA) / CFG 4, Step 50 (without accelerator LoRA)

Sampler: Euler + Simple

Denoise: 0.95

Recommended Resolution: 1920×1080 (native output works great)

Thank you for downloading and trying it out! If you enjoy my work, I'd really appreciate a follow. You can find me as chosen鲷 on both Bilibili and Pixiv—feel free to check out my channels:

Bilibili: https://space.bilibili.com/3867754?spm_id_from=333.337.0.0

Pixiv: https://www.pixiv.net/users/102907778

Thanks again! 😊