
✨ Gamy from Gokusotsu Kraken
— Gamy prowls like a spoiled queen, a sleek-cat smirk hiding the scratch beneath. She even talks about herself in the third person—“Gamy-sama”—a purr of arrogance wrapped in sugar. Coy one heartbeat, feral the next, she turns every encounter into a game of claws and curiosity. A glance, a grin, and suddenly everyone remembers who’s really running the room.
⚠️ WARNING
There are some inconsistent outfit design of her thorough the series. Sometimes it's sleeveless, sometimes it has single sleeves with bandana on her arm. This LoRA are optimized to presenting her in the sleeveless version.
🖼️ How I Generate My Images
🔄 UPDATED: This full breakdown covers my exact settings, prompting tricks, and how I survive with my “potato” GTX 1650 setup 😅
The article is kept current whenever my methods change, so you’ll always see my latest setup.
👉 Check it out here if you want to try the same method or copy my setup for your own gens.
🎨 Prompt Tips
These are optional starting points for handling a few behaviors of this version. You may still need some retries depending on the checkpoint, prompt, and LoRA strength.
🖤 Manga-like result
For a more manga-like result, avoid tags that strongly push anime coloring or an anime screenshot look in the positive prompt, also avoid any color tags. You also need to use :
✅ Positive prompt: greyscale
🖌️Anime Coloring Adjustment
If the result still keeps too much of the colored manga-page look, try adding
🚫 Negative prompt:textless_cover_page
This can help reduce the stronger manga-style color influence and push the result closer to anime-like coloring. Keep it as an optional helper tag rather than a required tag, since its effect may vary depending on the prompt and checkpoint.
🧥 Default outfit details appearing on other outfits
When a default outfit detail keeps appearing in a different outfit, add the exact leaking detail to the negative prompt.
For example, if the hairband keeps appearing when you do not want it:
🚫 Negative prompt: low_neckline
Use the specific leaking tag instead of adding many unrelated negative tags.
⚖️ LoRA strength and outfit flexibility
A weight of 1 can still work with a very different outfit, especially when the new outfit is clearly different from her default one.
Only lower the LoRA weight gradually when outfit details keep leaking even after using clear outfit tags and negative prompts. Lowering it too much may also weaken her face, hair, and overall character identity.
📊 Lora Project
If you want to know what characters that I'm covering next you can check these link :
👉 Personal Project
👉 Commision Project
If this model sparked something in you, show some love with a like, review, or buzz.
Now go—
Be wild
Description
This is the final version of this LoRA.
✅ Fixes from the Previous Version
More canon hair color
Her hair color is now closer to the original source material, thanks to the addition of colored images in the dataset.Corrected outfit
Her armwear and legwear have been adjusted to better match the canon design.More accurate facial details
The previous version was still usable, but this update improves several facial details to better match her original appearance.
🔧 Improvements
Reduced the LoRA file size to around half of the previous version.
Updated the dataset with better-curated and better-processed images.
Improved dataset tagging for better prompt control and outfit flexibility.
Improved accuracy for her default outfit.
Refined showcase prompts for more controlled results.













