banner_image ×
SeaArt AI Unternehmensversion
article cover

WAI-ANI-????-PONYXL User Guide

avatar
G
avatar_frame
GrayMan
veröffentlichte am Jan 13, 2025
70

WAI-ANI-????-PONYXL is one of the best anime focused checkpoints trained on Pony base model. Model's value comes from its high creativity about compositions and better prompt understanding skills. It's suitable for both basic and more complex prompts. Model has more stable and aesthetic anime style compared to standart Pony checkpoints.

SeaArt - Model Page

What This Model Excels At

  • Crafting high quality stylized anime arts.
  • `source_anime` tag is important for prompting, it can be used after score tags.
  • 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, also perform well both with simple sentence and tag based prompting.
  • Knowledge of many well-known anime/cartoon/game characters.
  • Knowledge of various face/body/characteristic/outfit terms.
  • Can create amazing background/landscapes.
  • More stable than other PonyV6 based models.
  • Great lighting/shadow usage over composition.

Cautions For Optimal Usage

  • Model uses negative prompts, but best usage is keeping it short and using recommended keywords.
  • Does not require quality modifiers like `masterpiece, best quality, hd` etc.
  • Make sure to use Clip Skip at `2`.
  • Model is suitable for LoRAs trained on Pony V6 model.
  • Doesn't require vae to perform well.
  • `score` tags must be used front of prompt.
  • Distance can reduce quality slightly but model still performs better than most of Pony models about long-distance renders.
  • Sometimes eyes can be rendered bad but using "Creative Upscale" fix this issue.

Recommended Settings

  • Steps: 30
  • CFG Scale: 7
  • Sampler: Euler a, DPM++ 2M Karras
  • Clip Skip: 2
  • VAE: None (User Preference)

Suggested Resources

LoRA

  • Model doesn't require any extra resource to perform well but LoRAs based on Pony models can be used.

Prompt Style

  • Requires `score_9, score_8_up, score_7_up` in front of positive prompt.
  • Keyword `source_anime` can be used after score tags to have more stable anime style.
  • `score_4, score_5, score_6` should be used in negative prompt for high quality outputs. Users also can prefer to use `source_cartoon` in negative to reduce style confusions.
  • Certain tags from Danbooru such as `lowres, monochrome, blurry, watermark, bad ?, bad hands, 3d, censor` also can be used in negative prompt to improve accuracy and quality of results.
  • Prefers tag like prompts and can understand most of Danbooru tags but also perform well with natural language prompts.

Recommended Prompt Order

  1. `score_9, score_8_up, score_7_up, source_anime`
  2. Subject
  3. Subject Details
  4. Pose/Action
  5. Additional Details of Composition (Background, theme etc.)
  6. Additional Modifiers and Tags

Prompt Examples

Protector of Abyss

Positive Prompt

`score_9, score_8_up, score_7_up, solo, 1girl \(protector of abyss, demigod\), upper body, white eyes, white hair, (white skin, pale skin), (floating hair:1.2), (black robe), casting spell, creating a (black hole:1.2), holding a black hole, dark magic, black background, abyss scenery, fallen stars, light particles, floating particles`

Negative Prompt

`score_4, score_5, score_6, lowres, worst quality, low detail, (blurry), watermark, 3d, (censor), child, childish, loli, ?, ?`

Farmer

Positive Prompt

`score_9, score_8_up, score_7_up, 1girl, farmer, pink hair, blue eyes, holding basket, carrying carrots, carrots in basket, walking on farm, wheat farm, barn, nature background, highly detailed, high resolution, 8k`

Negative Prompt

`score_4, score_5, score_6, lowres, worst quality, low detail, (blurry), watermark, 3d, (censor), child, childish, loli, ?, ?, deformed, jpeg artifacts`

You can join SeaArt Discord Community for more content, also you will have opportunity to choose next model guides with your vote. You can use comments or SeaArt server #guides-feedback to leave feedback or suggestion. We look forward to seeing you on the server.

Happy Generations To Everyone!

1515
70
0 Kommentar(e)
70
1.5K
0
0/400
Verwandte Tutorials
cover
Get Started with SeaArt SonoVision in 3 Minutes
avatar
S
avatar_frame
SeaArt Official
1665
1165
cover
COMPREHENSIVE GUIDE DIRECTORY
avatar
S
avatar_frame
SeaArt Official
11918
4873
cover
Wan2.5を使って10秒CMを作る(架空アパレルブランド編)
avatar
R
avatar_frame
razor
2468
360
cover
【Veo3キラー】Wan2.5 解説
avatar
椎
avatar_frame
椎名
1678
254
cover
Your Own World: How to Create Your Exclusive OC with SeaArt
avatar
S
SeaArt Guide Guy
1409
115
cover
【🍌】Nano Banana 解説
avatar
椎
avatar_frame
椎名
1408
385
cover
SeaArt Film Video Realistic Style Tutorial
avatar
C
chengxu01
283
8533
cover
Veo 3.1 is live on SeaArt!
avatar
S
avatar_frame
SeaArt Comfy Helper
272
93
cover
SeaArt Muse Prompt Tutorial
avatar
avatar_frame
SeaArt VIdeo Master
124
4319
cover
WANVideo VACE KJ Fun Use Case Collection
avatar
S
avatar_frame
SeaArt Comfy Helper
223
98
logo
Deutsch
Anwendungen
Bild erstellen KI-Charaktere Swift AI Modelltraining Canvas Quick Tool Arbeitsablauf
Infos
Studio Rangliste KI-Chat KI-Blog KI-Nachrichten
Hilfe
Anleitungen Kundenservice
APP hunterladen
icon
Download on the
APP Store
icon
GET IT ON
Google Play
Uns folgen
iconiconiconiconiconiconiconicon
© 2025 SeaArt, Inc.
Copyright Policy
Nutzungsbedingungen
Datenschutzrichtlinie 特定商取引法 資金決済法に基づく表示
Mehr