modèles/Amanatsu (Illustrious) - V1.1

Amanatsu (Illustrious) - V1.1

8/15/2025

4:11:55 AM

Mots-clés et tags associés

amanatsu,amanatsu (illustrious),anime,anime model,modèle de base,characters,checkpoint,detailed anime look,ikena,illustrious,nsfw,stable diffusion,v1.1,femme

A cute girl with brown eyes and double bun hairstyle wearing blue and white hanfu, sitting sleepily leaning against a tree in a detailed grassy landscape with mountains and trees in the background.

A surreal digital artwork featuring a winged female figure with flowing, colorful abstract patterns in red, blue, black, and white cascading from her dress and wings against a dark background.

Adult male with slicked back hair and sunglasses wearing a pink floral Hawaiian shirt, holding a glass of orange juice inside a cafe.

Anime-style female character Rem with blue hair and blue eyes wearing a black and white Roswaal mansion maid uniform in a dynamic pose

A cute cartoon girl with blonde hair and large sparkling eyes eating spaghetti with a fork in a warmly lit living room.

A mechanical maid with short white hair and a frilly black and white maid outfit dashes forward mopping the floor with a mop, showing a flustered expression. She has robotic arms and legs with white body panels accented in silver and blue, and a floating mini mecha with thrusters hovers behind her, all set against a clean white background with motion lines and cleaning bubbles.

Prompts recommandés

1girl, masterpiece

Prompts négatifs recommandés

signature

worst quality, bad quality

Paramètres recommandés

samplers

Euler a

steps

cfg

resolution

1024x1024, 1152x896, 896x1152, 1216x832, 832x1216, 1344x768, 768x1344, 1536x640, 640x1536, 816x1232, 960x1152, 824x1224, 896x1152, 880x1232, 832x1264, 800x1248

vae

sdxl_vaeFix.safetensors

Paramètres haute résolution recommandés

upscaler

R-ESRGAN 4x+ Anime6B, Latent (nearest-exact)

upscale

1.45 - 1.6

denoising strength

0.42 - 0.54

Conseils

Model is trained with images with minimal disruptive features like floating texts, logos, speak bubbles and signatures.

Use negative prompt 'signature' to avoid unwanted signatures in generated images.

Excluded metadata and franchise tags to improve output quality.

Points forts de la version

Update to Style B that is now amanatsu

Improvements:

- A bit more stable with backgrounds, less random strange extra stuff appears
- Less blurry
- Better eyes (if they don't artifact or used with hires or adetailer)
- Slightly less artifacs (still many, hires or adetailer are recommended)
- Brighter in most occasions, without loosing the ability for darker images

Its based on hassaku 2.1, but have back mergin with style V1. So its a bit of a combo of V1 and hassakuV2.1. The versatility of 2.1 + the style of B in one model.

Sponsors du créateur

My Discord for everything related to anime models and art. You can support me on my patreon.

Thanks to my supporters Riyu, SETI and Kodokuna on my patreon!

Amanatsu aims to be a model with a distinctive detailed anime look.

My Discord for everything related to anime models and art. You can support me on my patreon.

_____________________________________________________

Supporters:

Thanks to my supporters Riyu, SETI and Kodokuna on my patreon!

_____________________________________________________

Using the model:

Model is trained with images with minimal disruptive features like floating texts, logos, speak bubbles and signatures as possible. If such features appearing in a image, please use prompt "signature" as negative.
Excluded metadata and franchise tags, don't use them. meta data -> something like "highres", franchise tags like "re:zero kara hajimeru isekai seikatsu" are pruned on characters like "rem_(re:zero)".
Here are some good resolution options for SDXL:
- 1024 x 1024
- 1152 x 896
- 896 x 1152
- 1216 x 832
- 832 x 1216 (most recommended)
- 1344 x 768
- 768 x 1344
- 1536 x 640
- 640 x 1536