models/Hassaku (SD1.5) - V1.2

Hassaku (SD1.5) - V1.2

|
1/5/2026
|
5:46:43 PM
| Discussion|
0
Illustration of a mature female in a black kimono with a hair bun, sitting in seiza pose on a tatami floor indoors, with a gentle seductive smile.

Recommended Prompts

masterpiece,best quality

masterpiece, best quality, 1girl, nude, long hair

Recommended Negative Prompts

low quality,worst quality,monochrome,signature,text,logo

(worst quality, low quality:1.4)

Recommended Parameters

samplers

DDIM, DPM++ SDE Karras, DPM++ 2M Karras

steps

20

cfg

7 - 8

resolution

512x704, 512x600, 512x664, 512x768, 576x864, 432x800, 512x640, 576x704

Recommended Hires (High-Resolution) Parameters

upscaler

Latent (nearest-exact)

upscale

1.4 - 2

steps

20

denoising strength

0.48 - 0.55

Tips

Use mostly danbooru tags for prompting.

Use () instead of {} for weighting in prompts in stable-diffusion-webui.

Don't use face restore and avoid underscores in tags, e.g., type 'red eyes' not 'red_eyes'.

Don't go to really high resolutions because model quality drops and can cause artifacts.

For fixing black results (VAE bug) in web UI, use --no-half-vae in command line arguments.

Clip skip 2 is better for private parts, img2img, and prompt following, though clip 1 is visually better.

Version Highlights

Biggest change is a more drawn style with less extreme bright skin

What is better:

- Slightly more open pussy and anus (with highress fix)

- Anuses are less forgotten

- More responsive to nsfw tags

- Less realistic faces, without face defining prompts (ugly big lips faces, not good semirealistic)

- Slightly less chances of double limbs

What is better/worse for some:

- A little bit less "shiny", more drawn and less extreme bright skin (especially on highresfix)

- Clothes are more revealing

- Bigger possibility of motion lines, motion blur and floating symbols

What is worse:

- More chance of double anuses

- Less precise to create men

Hard to determine:
- Hands seems better then in 1.1, but hands are hands

- Highressfix seems more controllable (less changes on highress)


Trained with 30k NSFW images with later on fine-tune iterations. But needed heavy merging to make hands better again and reduce motion lines, motion blur and floating texts and hearts which are prominent in NSFW images. Problem with those, on highress, motion lines blend with the body and do spikes, motion blur extens to blurryness and floating texts/hearts morph into objects. The end result is much less changed as anticipated, sometimes identical, i decided to put it up as 1.2 version.

Creator Sponsors

Support me on patreon for early access to Hassaku and other models.

Join the Discord community for anime models and art.

Hassaku aims to be a model with a bright, clear anime style. Model focus are nsfw images, but also with a high emphasis for good looking sfw images as well. My Discord for everything related to anime models and art. You can support me on my patreon and if you are interested.

My models: sudachi(flat 2d), koji(2d), yuzu(light semirealistic), grapefruit

Supporters:

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

You can support me on my patreon, where you can get other models of me and early access to hassaku versions.

_____________________________________________________

Using the model:

Use mostly danbooru tags. No extra vae needed. For better promting on it, use this LINK or LINK. But instead of {}, use (), stable-diffusion-webui use (). Use "masterpiece" and "best quality" in positive, "worst quality" and "low quality" in negative.

My negative ones are: (low quality, worst quality:1.4) with extra monochrome, signature, text or logo when needed.

Use a clip skip 1 or 2. Clip 2 is better for private parts, img2img and prompt following. Clip 1 is visually better, because i assume, the model has more time and freedom there. I use clip2.

Don't use face restore and underscores _, type red eyes and not red_eyes.

Don't go to really high resolutions. Every model, like hassaku, get lost in the vastness of big images and has a much higher chance to greate, as example, a second anus.

_____________________________________________________

Loras:

Every LoRA that is build to function on anyV3 or orangeMixes, works on hassaku too. Some can be found here, here or on civit by lottalewds, Trauter, Your_computer, ekune or lykon.

_____________________________________________________

Black result fix (vae bug in web ui): Use --no-half-vae in your command line arguments

I use a Eta noise seed delta of 31337 or 0, with a clip skip of 2 for the example images. Model quality mostly proved with sampler DDIM and DPM++ SDE Karras. I love DDIM the most (because it is the fastest).

Previous
Phoenix | Sora SD1.5 - v1.0
Next
DucHaitenNiji - Cute_v2.3.5

Model Details

Model type

Checkpoint

Base model

SD 1.5

Model version

V1.2

Model hash

d3cd6ac55a

Creator

Discussion

Please log in to leave a comment.

Images by Hassaku (SD1.5) - V1.2

Illustration of a mature female in a black kimono with a hair bun, sitting in seiza pose on a tatami floor indoors, with a gentle seductive smile.

anime Images

Anime style elf girl with long silver twintails and green eyes, standing in a field of blue flowers, wearing a white capelet, striped shirt, and black pantyhose.
A young anime woman with blue eyes and brown hair stands before a cobalt blue mosaic floral background in a dreamscape aesthetic.
Anime style digital illustration showing a large, pointed triangular stone pyramid structure on barren land with scattered rocks under a star-filled sky and a planet with orange rings.
Detailed anime-style female warrior in black leather outfit, striking a dynamic pose with blue and orange colorful swirling effects in the background.
Anime character Frieren with pointy ears and green eyes, grey twintails, wearing a sheer white nightgown, sitting on a bed in a softly lit bedroom.
A detailed portrait of a cyborg with long white hair, blue eyes, and intricate robotic armor standing in a futuristic cyberpunk cityscape at night.
Close-up portrait of a petite elf girl with long silver twintails, teal eyes, and delicate facial features, wearing a white capelet with gold trim and red earrings, standing in a forest.
A delicate fairy resembling Tinkerbell with intricate translucent wings, sleeping peacefully on vibrant flowers, dressed in a detailed green outfit.
Y'shtola Rhul from Final Fantasy XIV wearing a black sleeveless Chinese dress with cat ears and a tail, set against a night background with hanging lanterns.
A futuristic sci-fi monastery and city built into steep cliffs of a lush gorge with a river below, featuring a flying machine and a solitary woman walking across a thin bridge.

base model Images

Photorealistic scene of undead characters including zombies and skeletons walking through a spooky cemetery illuminated by glowing jack-o'-lanterns under a dark, ominous sky.

hentai Images

Anime-style image of a young male character dressed in a maid outfit, created using Stable Diffusion
Anime-style boy with short black hair and a serious expression, wearing a maid outfit with a white apron and red bow tie, kneeling barefoot in a modern living room. AI generated image using stable diffusion.
An anime-style boy in a maid outfit with a red bow tie standing inside a modern living room with a camera on a tripod. AI generated image using Stable Diffusion.

nudity Images

A highly detailed AI generated image of a futuristic female cyborg with glowing blue accents, created using Stable Diffusion.

porn Images

Anime character in a dynamic pose wearing dark blue clothing with gold embroidery and rainbow gradient leggings, standing against a simple white background with a mystical blue wolf silhouette behind.
AI generated image using Stable Diffusion of a cyberpunk woman superhero with a bob haircut, wearing a futuristic red and black suit, showcased against a neon dark background with abstract text.
Anime female superhero standing in front of a futuristic cityscape. This is an AI generated image using Stable Diffusion.
A futuristic woman dressed in a red and black suit, standing under a full moon with glowing red dual swords. The painting has a cyberpunk aesthetic and is an AI generated image using Stable Diffusion.
Futuristic cyberpunk android with glowing eyes, intricate mechanical parts, and a vibrant color palette. AI generated image using Stable Diffusion.
A futuristic robot with neon accents and elongated limbs stretches beneath a vibrant blue sky. AI generated image using Stable Diffusion.
A futuristic cyborg-inspired female robot with red-and-black detail, created using stable diffusion AI technology.
A digital artwork featuring a human figure with intricate cyberpunk body art, generated using AI with Stable Diffusion.
A neon-colored AI-generated image of a futuristic woman with glowing elements in a cyberpunk style using stable diffusion.
A robotic shark's head with a glowing red eye seeming to emerge from a futuristic, monolithic building. AI generated image using Stable Diffusion.