This version has been trained with very descriptive captions and new data. It's very different from the previous models and prompts from previous models might not work.

Improved faces and bodies
Creates realistic, cinematic and cosplay images

There are no trigger words, just be descriptive in the prompt.

V3.1

Trained for another 200 epochs from V3. The extra training generates better faces and bodies but causes some burn in certain generations. If you encounter it, use face restore. There is a post in the last two images (woman in the yellow dress) that shows how to fix faces for full body poses as well.

Can create some cool sci fi generations as well. Aside from that, use just like V3.

V3

This version used BLIP2 captioning, and a mixture of nude and non-nude images with cinematic images in the mix. The following is different from V2:

Improves full body creations while sacrificing face fidelity. Use it with face restore.
Produces less NSFW compared to V2 unless prompted otherwise.
More mature faces.
Generates cinematic images.
Works well with LoRAs.
Can be used with higher CFG.

Sample prompts:

"a young woman, street, laughing, ponytails, (hdr:1.3), (muted colors:1.2), dramatic, complex background, cinematic, filmic, (artstation:0.8), soaking wet"

"photograph of a girl, street, <lora:A-Mecha-REN[color_theme,mecha musume, mechanical parts, robot joints,headgear,full armor]:1>,black_theme,mecha musume, mechanical parts, robot joints,headgear,full armor,ponytails, (hdr:1.3), dramatic, complex background, cinematic, filmic, <lora:GlowingRunesAIV6:0.8>, GlowingRunesAIV2_red"

"photograph of a woman, street, skirtlift, shirtlift,<lora:skirtlift-v1:1>, (hdr:1.3), (muted colors:1.2), dramatic, complex background, cinematic, filmic, (artstation:0.8), soaking wet, <lora:shirtliftv1:1>"

Negative prompt:

"child, deformed, deformed face, unreal, bad hands, bad fingers, monochrome, over saturated, bad hands, lowers, 3d render, cinematic, cartoon, long body, blurry, duplicate, duplicate body parts, disfigured, poorly drawn, extra limbs, fused fingers, extra fingers, twisted, malformed hands, mutated hands and fingers, contorted, conjoined, missing limbs, logo, signature, text, words, low res, boring, mutated, artifacts, bad art, gross, ugly, poor quality, (low quality:1.2), (worst quality:2)"

"(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, tattoo, extra heads, art, unreal"

Scale:

5-10

Steps:

20 - 50

clip_skip: 2

V2

Very NSFW even with negative prompt added keywords "nude" and "naked", but can also create SFW.

Sample prompts:

"close up of a black girl smiling looking at the camera"

"highly detailed photography, spring, portrait of beautiful blonde woman standing in a park, top and miniskirt, perfect face, perfect body, large breasts, slightly thin skin, sharp focus, photo by Annie Leibovitz, intricate, natural lighting, high quality, 4k, cover photo"

"a 25 year old European girl with red hair wearing a white bikini top and denim skirt, posing for the camera by the beach, in a boat"

Negative prompt :

"nude, naked, child, deformed, deformed face, unreal, bad hands, bad fingers, monochrome, oversaturated, bad hands, lowers, 3d render, cinematic, cartoon, long body, blurry, duplicate, duplicate body parts, disfigured, poorly drawn, extra limbs, fused fingers, extra fingers, twisted, malformed hands, mutated hands and fingers, contorted, conjoined, missing limbs, logo, signature, text, words, low res, boring, mutated, artifacts, bad art, gross, ugly, poor quality, low quality"

Scale:

3-5

Steps:

50 - 100

clip_skip: 2-4