Amanatsu aims to be a model with a distinctive detailed anime look.
My Discord for everything related to anime models and art.
If you'd like to support my work, you can do so at: SubscribeStar, every bit helps and is truly appreciated!
____________________________________________________________
Supporters:
Thanks to my supporters tamashiicolle, pttcot, SETI and Kodokuna
_____________________________________________________
Using the model:
Model is trained with images with minimal disruptive features like floating texts, logos, speak bubbles and signatures as possible. If such features appearing in a image, please use prompt "signature" as negative.
Excluded metadata and franchise tags, don't use them. meta data -> something like "highres", franchise tags like "re:zero kara hajimeru isekai seikatsu" are pruned on characters like "rem_(re:zero)".
Here are some good resolution options for SDXL:
1024 x 1024
1152 x 896
896 x 1152
1216 x 832
832 x 1216 (most recommended)
1344 x 768
768 x 1344
1536 x 640
640 x 1536
Description
Amanatsu 1.1 is a bit older now and have a difference of 9-months training to V2 and needed another set of style merges to work, I tried to keep it as close as possible to the original. As a result, some concepts may look quite different in this update.
The biggest difference is from the new base model acquired sensitivity to artist tags. If you used artist tags before, you will notice that the results can look completely different to version 1.1.
FAQ
Comments (17)
I hope this gets an update soon. It ain't broke, but hassaku getting all the love is disappointing..
you can join my discord and watch more or less me working on a update on amanatsu in my what i do channel talk https://discord.com/channels/1072156418402156624/1425222621955625250
my favorite checkpoint, bar none
oh, unexpected
Amanatsu my beloved. Come home 😭
i hope 1.1 comes back :( the v2 looks nice but it isn't the same style at all.
Just download it
Can't wait for the V2 to be free, this model in V1.1 is very underrated, very good style, the base lighting and base detail without lora is really good compared to something like even WAI, maybe it's just my skill in prompting, but this model still going to be my favorite
v.2 is much, much worse than v.1 or v.1.1 Style is really different, to the garbage side. Dont use 2.0, while 1.0 one of the best models I'd ever use
What's worse? The style or are there other issues?
@ClownReplica Well, both, actually. Style is missplaced and artists tag are not reading normally. Also, much more artifacts especially with eyes
Based on my testing, version 2.0 doesn't have a worse style; rather, it's more sensitive to style prompts. Therefore, the weights of each prompt need to be readjusted.
Even after updating to version 2, the variety of images displayed for the same prompt remains the same and far exceeds other models. However, the art style has changed.
This is one of the few models that knowing of aglaea \(honkai: star rail\) appeared in February 2025.
There are many other models, but don't knowing the new character means that just created a cocktail by merging with other models rather than learning the model.
However, this model proved that creator did effort extra learning.
I deleted every models that don't know aglaea. because making just a cocktail is not only easy but also burying real creators.
thank you very much. and keep go on.
There are so many models now, "updated" every few weeks. But... many of them dont feel carefully crafted at all, they feel like slop and generations outside 1girl are a inconsistent mess!
Many times I just end up using good old Amanatsu! All versions are great in my opinion.
It’s not quite as pretty as 1.1 out of the box, but it reacts to Lora differently. 1.0–1.1 are legendary, but this one isn’t bad at all. Thanks for the great work.
Hassaku結構好きなんですが、Amanatsuの方が良い感じになることが多く、最近はこっちの割合が多いです。こっちはバージョンアップ予定はありますか？