T-Ponynai3 V6

T-ponynai3 V6 is a model tailored for generating distinct anime art with vivid colors, utilizing built-in features to simplify usage and improve style accuracy.

T-ponynai3 is an high quality pony based model. It specializes at generating anime art with certain style. Uses negative prompts and works differently from other pony based models. Model already has included vae, so you don't have to use "sdxl_vae" as other models. Term "anime" is used on training model, which means using "anime" in your prompt will reduce style confusion and model will focus on certain anime style instead of western style cartoons. Style of model is different from general anime models and looks pretty with vivid colour selection

Try the Model: T-Ponynai3 V6

What This Model Excels At

• Crafting stylized anime arts.

• Has Pony Diffusion based special tags for better experience. Only source_anime and rating tags are recommended because model focuses on anime arts.

• Generating human/humanoid, however model still can generate pony/furry images but it's not main purpose of model. (Not recommended to use for furry/pony images.)

• Generating female/male/transgender subjects.

• Crafting SFW/NSFW content imagery.

• Crafting images with wide-range themes. (fantasy, sci-fi, medieval etc.)

• Recognizes many of Danbooru based tags, but model training is not Danbooru based.

• Recognizes only artist database of Pony Diffusion, no artist added from nai3.

• Knowledge of many well-known anime/cartoon/game characters. (Especially Genshin Impact characters)

• Knowledge of various face/body/characteristic/outfit terms.

• Can create amazing background/landscapes. (Especially V6)

• More stable than other PonyV6 based models.

• Better at hand, feet renders.

• Great lighting/shadow usage.

Cautions For Optimal Usage

• Model uses negative prompts, but best usage is keeping it short and using recommended keywords.

• Keyword anime can be used to increase style of image.

• Does not require quality modifiers like masterpiece, best quality, hd etc.

• Make sure to use Clip Skip at 2.

• Perfectly supports LoRAs trained with PonyV6 as the base model, yet AnimagineV3 and SDXL1.0 LoRAs can be used sometimes but not recommended.

• Do not use sdxl_vae, model already has included vae.

• Can't create photorealistic images but pretty good at rendering realistic cartoon images.

• Use anime (for nai3) and source_anime (for PonyV6) tag to achieve desired anime style.

• score tags must be used front of prompt.

• 1024px resolution is recommended, although model can generally work with most of supported SDXL resolutions.

Recommended Settings

• Steps: 25-30

• CFG Scale: 7

• Sampler: Euler a

• Clip Skip: 2

• VAE: None, already includes vae

Suggested Resources

LoRA

• LoRAs based on Pony Diffusion V6 XL

• SDXL style LoRAs (can cause error/crash)

Prompt Style

• Requires score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up or score_9, score_8_up, score_7_up in front of positive prompt.

Note: Full string should be more effective, try and select one you prefer.

• Model uses negative prompts, unlike the other PonyV6 based models. (score_4, score_3, score_2, score_1) Model uses lower score tags at negative while many Pony models don't have them.

• Use ugly, bad anatomy, bad hands, bad feet like terms in negative to improve accuracy of generation.

• Keyword anime and source_anime can be used to acquire desired style.

• Data selection tags can be used to make AI focus on certain contents.

• Can work with both of short and long prompts.

• Prefers tag like prompts. Model isn't Danbooru based but can understand Danbooru tags.

Recommended Prompt Order

1. score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up

2. source_anime, anime & Data Selection Tags (User Preference)

3. Subject

4. Describe your suggested composition in general order. Use tag based prompts for better understanding.

Special Data Selection Tags

• Tags can be used in both positive and negative prompts to focus on a specific data of the Model's training database.

Source

• Model excels in anime style, be sure to use anime (for nai3) and source_anime (for PonyV6) in your prompt. You can use other source tags too.

• source_cartoon (Not Recommended)

• source_pony (Not Recommended)

• source_furry (Not Recommended)

Content Safety Rating

• rating_safe

• rating_questionable

• rating_explicit

Prompt Examples

Raiden Shogun

Positive Prompt

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, (anime), (front view), solo focus, 1girl, raiden shogun, [disdain|haunted] face, mole under eye, holding, holding weapon, holding sword, (sword between breasts), (human scabbard:1.2), electricity, musou isshin (genshin impact), sakura forest, (purple sky:1.2), [[lightning]], aesthetic pose, ready-to-fight

Negative Prompt

(score_4, score_3, score_2, score_1), ugly, bad anatomy, bad hands, (bad composition, wrong composition)

LoRA

• Booba Sword (Human Scabbard)

Girl's Abyss

Positive Prompt

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, solo focus, sfw, ((portrait of a girl)), upper body, abstract art, surreal, touching a butterfly, floating hair, blue butterfly, glowing scenery, (falling petals, rose petals:1.2), blue theme, crimson theme, (aquamarine theme:1.2), (dark contrast:1.3), color chiaroscuro, (abyss, mysterious fields:1.3), amazing background, subsurface scattering, whimsical ethereal art

Negative Prompt

(score_4, score_3, score_2, score_1), ugly, bad anatomy, bad hands, nsfw