illust-Anime_IL

CHECKPOINT
Original - TenStar Fund


Updated:

niji_v4: this version is closer to V2 and V1, smoother than V2 and less complex, making it easier to use Lora. It is still closer to illustration, animation, and thick painting. The usage method is the same as V2, and DMD2 acceleration is not added. I hope you like it.

niji_v3: 2.5D, Anime, PaintNiji, Men's Specialty.

I was originally planning to start a new series, but since it's similar to another new checkpoint, I'm posting it here instead. So, it's quite different from V2 and V1. Combined with my Niji Lora, the effect is astounding. Please try it for yourself. Have fun!

niji_V2: 2D, 2.5D, Semi-realistic, Anime, Paintniji, Smoothniji, stable body structure and posture, diverse composition, no AD required.

For a mild Niji style, if you want the Junkjuice effect, you can add Loras created by various experts. Experiment more, and enjoy your time!

Euler/Euler a+karras:Anime

dpmpp_sde_gpu/dpmpp2m_sde_gpu+karras:Paintniji

dpmpp2/dpmpp2m+karras:Smoothniji

Steps: 20-40, CFG scale: 4-7, Sampler: Euler a/dpmpp_sde/dpmpp2m_sde, scheduler: karras/sgm_uniform/exponential/beat

The VAE is already integrated

use size larger than1024x1024 for the original dimensions.

Positive Prompt

masterpiece,highres,best quality, very aesthetic,

Negative Prompt

bad quality,worst quality,worst detail,

If you think the details are not enough, you can refer to the lora I often use. If you are not satisfied with the face.You can try these artists (can be used in combination):

female: (artist:quasarcake),(artist:moriimee),(artist:honjou raita),(artist:lack),(artist:rella),(artist:wanke),(artist:748cm),(artist:cutesexyrobutts),(artist:Rinotuna),

male:(artist:remsrar), (artist:mazjojo), (artist:pigeon666), (artist:zawar379),

My usual settings are: 25, 5, dpmpp_sde_gpu, karras, 832x1536,

Prompt words and Danbooru tags are the key to making the most of all Illustrious and NOOB features. Everyone, give it a try. Have a great time!

Version Detail

Illustrious
This version is closer to V2 and V1, smoother than V2 and less complex, making it easier to use Lora. It is still closer to illustration, animation, and thick painting. The usage method is the same as V2, and DMD2 acceleration is not added. I hope you like it. Euler/Euler a+karras: Anime dpmpp_sde_gpu/dpmpp2m_sde_gpu+karras: Paintniji dpmpp2/dpmpp2m+karras: Smoothniji Steps: 20-40, CFG scale: 4-7, Sampler: Euler a/dpmpp_sde/dpmpp2m_sde, Scheduler: karras/sgm_uniform/exponential/beat The VAE is already integrated. Use a size larger than 1024x1024 for the original dimensions. V3 is the result of my upgrade to Illustrious 2.0Pro. I may open a new branch to continue updating it in the future. If you like V3, please give me feedback. I will not update a version with such a significant difference in quality in the future.

Project Permissions

Reprinting is strictly prohibited

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Related Posts