■This is an experimental ground for Anima LoRAs.
The LoRAs I create serve more of a role in stylistic tuning rather than reinforcing single concepts.There is no inherent superiority among the LoRAs; they simply serve different purposes. You can find the description for each LoRA in its respective tab.
■If you notice pixel patterns in your generated images, try the VAE below. It removes the pixelation typical of the qwen_vae and makes the images much sharper.
https://civarchive.com/models/2718533/hdr-vae-anima-qwen-image
■If the VAE looks a bit too saturated for your taste, try using this VAE merge node to blend it with the original qwen_vae to dial it in. A 0.5 merge should be just right.
https://civarchive.com/models/2487550/vae-merger-for-comfyui-anima
■I would like to share the possibilities of anima with everyone.
My wish is for many people to discover basemodels with potential and to see their possibilities unfold even further. I would be happy if I can help make that happen.
■The anima architecture has been heavily modified and retrained from scratch—almost like a radical overhaul—but the base architecture is Cosmos-Predict2-2B. To put it simply, you could compare it to a lightweight version of Flux 1.
I think it’s a great architecture capable of generating high-quality images despite its small size.
I have a very good impression of Anima when it comes to color vibrancy, lighting, and prompt adherence.
Its knowledge regarding NSFW concepts, characters, and styles is at least on par with SDXL. The overall feel of using it is also quite similar to SDXL.
Furthermore, it is lightweight and easy to train. I believe large-scale training is also a viable option.
Since we are provided with a base model that already contains a vast amount of knowledge, we rarely need to teach the model completely unknown concepts from scratch.
Basically, our work mostly comes down to adjusting styles or reinforcing minor concepts and characters. You can achieve great results without having to do any heavy lifting.
The developer has also ensured transparency by sharing the training settings, which means you are less likely to stumble over black-box issues.
Because of this, the burden of training is low, making it an option well worth nurturing together as a community.
■The LoRA has a stronger influence when used with shorter prompts.
Since longer prompts often carry their own inherent style, the LoRA will primarily take on a supporting role to refine and unify the overall image.
■https://github.com/gazingstars123/Anima-Standalone-Trainer
This is a great tool that allows for easy training, even on Windows. It has everything you need for LoRA training.
Even if you run into any issues, you should be able to get your LoRA training up and running by troubleshooting with ChatGPT or Gemini.
If you find the tool useful, you might want to consider making a donation to him. It allows the developer to focus more on development and deliver even better tools, which ultimately benefits you in the long run.
■If you have any questions, please feel free to ask!
日本語での質問も大丈夫ですのでご気軽にお声がけください～
If you enjoy my work and feel like making a donation, I would be incredibly grateful. Even the smallest token of support goes a long way in helping me lighten the load.
Whether I receive donations or not, I will keep doing what I love and pursuing my interests just as I always have. However, simply knowing that this kind of support exists brings me immense joy and serves as a huge driving force for me.
Description
This is a LoRA for the masterpiece movie The Vision of Escaflowne, which stands as the absolute pinnacle of cel animation.
■Dataset & Training This model was trained on a carefully curated dataset of 1,150 images. In addition to the art style, it has also learned specific characters (the character list is provided at the bottom of this page).
■Style Characteristics The overall aesthetic features distinctive, almost 2.5D-like facial structures, high-contrast lighting, and a palette of rich earth tones. It also beautifully captures the unique post-processing effects inherent to classic cel animation.
The style trigger is "tenkuu no escaflowne".
■Prompting (Natural Language vs. Tags) For this model, I utilized natural language captioning during training (captions were generated via Gemini). While standard Booru-style tag prompting works perfectly fine as always, you might find that natural language prompting yields smoother results. If you want to try the natural language approach, I highly recommend finding a reference image close to what you want to generate and having Gemini write a descriptive caption for it.
■A Tribute to 90s Sunrise Animation In the 1990s, Sunrise produced numerous masterpieces—Gundam 0083: Stardust Memory, The 08th MS Team, Cowboy Bebop, and more. Each of these is a masterclass in cel animation that reaches the realm of fine art. Escaflowne can truly be called the culmination of that golden era. Every single cut looks like a painting, with lighting and color design that are simply in a league of their own. Standing shoulder-to-shoulder with Vampire Hunter D, Escaflowne represents the ultimate peak of cel animation. Created right at the turning point as the cel animation era came to a close, this film is packed with everything that makes the medium so magical. If you haven't seen it yet, it is absolutely worth your time.
■ Character List Please note that characters with limited appearances in the dataset might have lower reproducibility (generation accuracy). Also, the character descriptions provided are just rough summaries.
kanzaki hitomi: The main protagonist (The girl from Earth).
van fanel: The main protagonist (The young king of Fanelia).
allen schezar: The blonde-haired leader of the "Abaharaki" rebel group.
millerna aston: A pink-haired female member of the Abaharaki.
merle: A young cat-girl (feline beastman).
jajuka: A kind-hearted dog-beastman.
folken fanel: Van's older brother.
mole man: A mole beastman and fortune teller.
dilandau albatou: A white-haired young man with a tragic background of being bullied and biologically modified.
siesta: A clairvoyant character who stays close to Dilandau.
sora: A woman who serves closely by Folken's side.
eriya: A beast-woman tavern singer featuring dark skin and blonde hair.
naria: A beast-woman tavern singer featuring pale skin and silver/white hair.
dryden Fassa: The owner of the tavern.
nukushi: A good-natured beastman who is tragically killed by Dilandau.
gades: A knife-throwing member of the Abaharaki.
uchida yukari: Hitomi's friend from Earth.
escaflowne \(guymelef\): The titular main mecha.
alseides \(guymelef\): The enemy mecha.
escaflowne \(dragon mode\) \(guymelef\): The dragon transformation of Escaflowne.
escaflowne \(dark mode\) \(guymelef\): The corrupted form of Escaflowne that brings about the world's destruction.