Use it respectfully&gt; 👍 Leave a Like — I'm just getting started and it would mean a lot! Have fun! 🔥 🔥 Bakugo Katsuki | LoRA — My Hero Academia | Base: ANIMA📦 Trained with 265 diverse images. Available in two versions, both trained with the same steps.---&gt; ⚠️ LoRA Strength tip: At strength 1.0 the model will almost always default to the hero costume. For more outfit flexibility, try values between 0.6 – 0.7. Above 0.8 it becomes harder to change clothing, though the overall quality tends to be better. At 1.0 you can still swap outfits, but it requires a more detailed prompt.&gt; ⚠️ The model only uses images of Bakugo — Midoriya might not come out as expected. Use both LoRAs together if you want better results with the two characters.---⚙️ Generation settings used for preview imagesAll preview images were generated using:- Base model: Anima 1.0- Acceleration LoRA: strength 1.0- CLIP: anime default- Steps: 12- CFG: 1.0- Sampler: <code>er_sde - beta</code>---## ⚡ Rank 24Slightly softer linework and style. Handles multi-character scenes better. Tends to enforce the hero costume more consistently, making it easier to get him in uniform without heavy prompting.---## 💥 Rank 32More intense and muscular interpretation . A bit more flexible with outfit variation, giving you more room to experiment with different clothing. May require a bit more prompting to keep the hero suit if that's your goal.---Both versions share the same training steps, so the difference comes down to style preference and use case.Compared to the vanilla ANIMA base model, the improvement in Bakugo's likeness is very noticeable — face accuracy, expression, and overall style are significantly closer to the source material.

Katsuki Bakugo MHA

Bakugo-step00001400.safetensors

RANK 32

RANK 24

<h2 id="bakugo-katsuki-or-lora-my-hero-academia-or-base:-anima">Bakugo Katsuki | LoRA — My Hero Academia | Base: ANIMA</h2>🙏 Before anything else: this LoRA took a lot of work — picking the images, building the dataset and tags, training, and 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. 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 265 diverse images. Available in two versions, both trained with the same steps.<hr /><h2 id="about-this-model">📊 About this model</h2>Compared to the vanilla ANIMA base model, the improvement in Bakugo's likeness is very noticeable — face accuracy, expression, and overall style are significantly closer to the source material.⚠️ LoRA Strength tip: At strength 1.0 the model will almost always default to the hero costume. For more outfit flexibility, try values between 0.6–0.7. Above 0.8 it becomes harder to change clothing, though the overall quality tends to be better. At 1.0 you can still swap outfits, but it requires a more detailed prompt.⚠️ The model only uses images of Bakugo — Midoriya might not come out as expected. Use both LoRAs together if you want better results with the two characters.<hr /><h2 id="generation-tips">ℹ️ Generation tips</h2><ul><li>On ANIMA base with 12 steps + acceleration LoRA, the output keeps a clean anime feel that works well for Bakugo — recommended as a starting point</li><li>Using more steps or a different ANIMA-based model with quality LoRAs will improve sharpness and detail</li><li>Higher CFG can help lock in his facial features and outfit details more consistently</li><li>Check the prompt examples used in the posted preview images — they show how to structure tags for each outfit</li></ul><hr /><h2 id="versions">📊 Versions</h2><h3 id="rank-24">⚡ Rank 24</h3>Slightly softer linework and style. Handles multi-character scenes better. Tends to enforce the hero costume more consistently, making it easier to get him in uniform without heavy prompting.<h3 id="rank-32">💥 Rank 32</h3>More intense and muscular interpretation. A bit more flexible with outfit variation, giving you more room to experiment with different clothing. May require a bit more prompting to keep the hero suit if that's your goal.Both versions share the same training steps, so the difference comes down to style preference and use case.<hr /><h2 id="generation-settings-used-for-preview-images">⚙️ Generation settings used for preview images</h2>All preview images were generated using:<ul><li>Base model: Anima 1.0</li><li>Acceleration LoRA: strength 1.0</li><li>CLIP: anima default</li><li>Steps: 12</li><li>CFG: 1.0</li><li>Sampler: <code>er_sde - beta</code></li></ul><hr /><h2 id="one-more-thing">👍 One more thing</h2>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! 🔥

Files with SHA256: fbb2312c0d7ab7a75291ace335c49656528f6b8365dd2abfd116e242b1a551b0