Updated: Sep 6, 2026
styleDownload
1 variant available
fp32 SafeTensor
YOURNAME_H3_V1-step00009900.safetensors
Full precision, largest file • 284.44 MB
Verified: 2 days ago
3950 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
10 1 2 3 4 5 6 7 8 9
(36)
Sep 6, 2026
MiniMax H3
Trained on 228 screen captures from Makoto Shinkai's "Your Name". The dataset is entirely environmental shots, with a few people and very few close ups of people if any. Trained on 1536x864 resolution.
Trigger word:
2d anime style, art by makoto shinkai
The base model is definitely trained on this movie, and probably a bunch of his other works. But I wanted to see if I could get more detail out of the environment. And I think it did sort of work in a way as a detailer. Things look more Shinkai rather than Ghibli now. You get a lot of the camera composition from shinkai as well, the lense flairs, the bokeh without mentioning them in the prompt. How environmental items are placed in the foreground. How the lighting works in animation etc. All that seems burned into the lora, which I think you could get out of the base model but this is more consistent. Character design though did find its way, there is only so many shots where people were in the dataset, but I think that's fine.
Show more

2.5K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
3K0 1 2 3 4 5 6 7 8 9K
MiniMax H3 is licensed by MiniMax under the MiniMax H3 Community License Agreement. That agreement’s Applicable Territory excludes the European Union, the United Kingdom, the Republic of Korea and the United States of America. Your use of H3 and of any H3 derivative is subject to that agreement and its Acceptable Use Policy.
MiniMax H3
H3:
Trained on 228 screen captures from Makoto Shinkai's "Your Name". The dataset is entirely environmental shots, with a few people and very few close ups of people if any. Trained on 1536x864 resolution.
Trigger word:
2d anime style, art by makoto shinkai
The base model is definitely trained on this movie, and probably a bunch of his other works. But I wanted to see if I could get more detail out of the environment. And I think it did sort of work in a way as a detailer. Things look more Shinkai rather than Ghibli now. You get a lot of the camera composition from shinkai as well, the lense flairs, the bokeh without mentioning them in the prompt. How environmental items are placed in the foreground. How the lighting works in animation etc. All that seems burned into the lora, which I think you could get out of the base model but this is more consistent. Character design though did find its way, there is only so many shots where people were in the dataset, but I think that's fine.
I almost didn't upload this, because there isn't huge changes from the base model, mostly subtle. But I do like certain aspects, so I thought I'd share. Let me know what you think.
LTX:
Trigger words:
ANIMSTY, SHW_YN.
I took my 90's anime lora at 53K steps and trained on top of it 231 clips from Makoto Shinkai's "Your Name" an additional 11.5k steps. But all the dataset was only environmental shots. Landscapes, rooms, cityscapes, etc. basically anything not human focused. There are some crowd shots and some scenes do have people though. As a result the people are more in the 90s anime style design but the animation look and feel is more modern. The downside is that the model started to trend back towards a more "3D" look and motion on the people.
This was mostly a test to see if a detailed work like Shinkai's could act a detailer for environments without too much negative effect on other aspects. And I think for the most part it worked, even the people's faces have more details. I think this dataset will work well mixed into a larger pool for a modern anime lora. Give it a try and let me know what you think.

