Eijiro Kirishima (Red Riot) | LoRA — Base: ANIMA

🙏 Before anything else: this LoRA took a lot of work — about 5h picking the images, 4h building the dataset and tags, 5h of training, and another 2h testing the checkpoints. If you have a moment, drop a Like and especially post your results in the comments (with the prompt/settings you used) — it helps me a lot to know what to improve in the next versions. 🔥

⚠️ The model was trained without any NSFW images. Will it generate NSFW content? Use the adult version of the character. Use it respectfully.

💬 Check out how the prompt examples were set up in the posted images.

💬 Leave your generations in the comments, and tag the LoRA so I can see how it's looking, it would help a lot.

📦 Trained with ~566 images, with at least ~546 sourced directly from the anime. Releasing the 1000 steps checkpoint as it gave the best results — other checkpoints up to 2000 steps available on request.

📊 About this model

On the vanilla ANIMA base, Kirishima tends to lose his sharp teeth, spiky red hair shape, and the details of his Hero Costume armor — he comes out as a generic red-haired anime boy instead of Red Riot. This LoRA fixes that: it locks in his hairstyle, sharp teeth, eye color, and outfit details (Hero Costume, school uniform, blazer, gym uniform, and more) much more consistently across different poses and outfits.

ℹ️ Generation tips

• The preview images were generated at 1024x1536 (vertical), which helps a lot with full-body framing — but you can also generate in other proportions. You just need to make sure the resolution is based on 1024 (for example, 1024x1024, 1024x1344, 1024x1536) to get the best results, but nothing stops you from using smaller resolutions; anything bigger than 2x might not turn out as well.

• On ANIMA base with 12 steps + acceleration LoRA, the output keeps a clean anime feel that works well for Kirishima — recommended as a starting point

• Using more steps or a different ANIMA-based model with quality LoRAs will improve sharpness and detail

• Higher CFG can help lock in his facial features and armor details more consistently

• Check the prompt examples used in the posted preview images — they show how to structure tags for each outfit

📊 Checkpoints

StepsResult⭐ 1000Best balance of style and likeness — recommended, the one being released

Up to 2000 Available on request — slightly more "locked in" details, but composition/pose gets more repetitive past 1000-1200

Want a different checkpoint? Just ask in the comments — I have steps available up to 2000 and can upload them on request.

⚙️ Generation settings used for preview images

All preview images were generated using:

• Base model: Anima 1.0

• Acceleration LoRA: strength 1.0

• CLIP: anima default

• Steps: 12

• CFG: 1.0

• Sampler: er_sde - beta

• Resolution: 1024x1536

👍 One more thing

I'm just getting started with this — if you have a moment, drop a Like and share your generations with the settings you used. It really helps me improve the next versions. Have fun! 🔥