Sign In

Typhoeus 提丰 — Arknights: Endfield 明日方舟:终末地 [Anima+Illustrious]

108

Download

1 variant available

fp16 SafeTensor

typhoeus_anima_v1.safetensors

Half precision, best balance • 87.61 MB

Verified:

Type
LoRA
Stats

175

5

2

Generation License Fee

5 / 10 images

Reviews
Published

Aug 22, 2026

Base Model

Anima

Hash
AutoV2
DD3902C818
Trigger Words
typhoeusendfield
default creator card background decoration
Followers - 368

368

Likes - 2005

2K

License:

Anima

The Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Built on NVIDIA Cosmos

💬 Discord: join here · ☕ Ko-fi: support the factory

Typhoeus 提丰 — Arknights: Endfield 明日方舟:终末地

Horns, drills and a tail. All of it rides the trigger.


Character LoRA for Typhoeus, trained with a baked identity. The trigger alone carries the canonical look: the lavender twin-drill hair, the pink eyes, the curved black horns and their headgear, and her white-and-black rig with the red emblem. None of it needs prompting.

Her outfit pieces stay tagged, so they remain steerable. That cuts both ways — name them when you want them exact, drop them when you want her in something else.

📋 Prompt

Default look. The trigger on its own already gives the full canon rig:

typhoeusendfield, 1girl, solo, cowboy shot, looking at viewer

Pinned canon kit, when you want every piece exact:

typhoeusendfield, 1girl, solo, horn hairband, white thighhighs, detached sleeves, black gloves, bare shoulders, tail

With her bow:

holding bow (weapon), weapon

Negative, worth keeping:

chibi, speech bubble, english text, multiple views, watermark, signature, meme

NO quality tags needed. Keep the rest lean, long tag stacks dilute the trigger.

🧩 Version

  • 🔵 v1.0 Anima, for Anima / Qwen-Image DiT checkpoints. Weight 1.0 (usable range 0.85 to 1.0).

  • 🟢 v1.0 Illustrious, for SDXL / Illustrious / Pony / NoobAI checkpoints (showcases rendered on WAI). Weight 1.0 (usable range 0.85 to 1.0). Same trigger, same prompt stacks.

💡 Tips

  • Small, dense dataset. This is a 16-image bake — the identity is strong, but she has fewer poses and angles in her than a big LoRA. If a pose fights you, say it explicitly rather than rerolling.

  • Name the facing. Scene prompts without looking at viewer will sometimes turn her around — several training shots are back or three-quarter views. facing viewer settles it.

  • Never negative her colours. purple hair, pink eyes and horns were stripped from every training caption, which is why they ride the trigger. Negating them fights the LoRA.

  • The tail is canon kit, not a bonus — tail keeps it in frame on full-body shots.

  • Alt outfits are easy. The clothes are tags, not the trigger. Dress her in anything and the horns, hair and face hold.

  • Keep solo in. A few training images are group shots, so without it company turns up.


More of my LoRAs: Techannel on Civitai. Likes and Buzz keep the factory running ❤️