Model
Trends
.ai
Model
Trends
.ai
de bästa AI-modellerna med öppen källkod

WAI-illustrious-SDXL v16: character-rich anime with simple prompts

Ämne: Modeller
Av Captain
Publicerad 2026-10-03
WAI0731's WAI-illustrious is one of the most used Illustrious checkpoints, with knowledge of thousands of characters. Learn its light-prompt philosophy, settings, aging tags and ideas for striking anime illustrations.
Den här sidan har inte översatts ännu, så den visas på engelska.

Översikt

WAI-illustrious-SDXL by WAI0731 is an Illustrious-based anime
checkpoint
known for recognizing thousands of characters by name and for producing clean, vivid illustrations with very short prompts. Version 16 refines color consistency and background relevance.
The author's main advice runs against habit: do not pile on quality tags or long
negative prompts
, because they make the image blurrier. A hires fix pass is recommended and often repairs limbs on its own.

Referens

Namn
Typ
Vad det är
Prompt
tags
A few Danbooru tags plus one or two quality tags. Short is better.
Negative prompt
tags
Keep it to a handful of words: lowres, bad anatomy, worst quality.
CFG
5-7
6 is a good default.
Steps
25-30
Euler a.
Hires fix
1.5x, 0.4-0.5 denoise
Recommended; it fixes arms, legs and hands frequently.
AD
Inget att installera - kör AI-modeller i molnet
BitVector Prism är det enklaste sättet att komma igång: välj en modell, skriv en prompt och generera i en ren webbapp, utan någon installation alls. BitVector finns också på Discord och på webben (SpyGlass).

Steg för steg

  1. Write masterpiece, best quality, then the subject and setting tags.
  2. Name a character or describe one with a few appearance tags.
  3. Use a short negative prompt and
    CFG
    6, Euler a, 28 steps.
  4. Enable hires fix at 1.5x with 0.45 denoise.
  5. If the character looks too young for the scene, add (aged up:1.2) or (mature female:1.2) at the start of the prompt, as the author suggests.

Exempel

Minimal prompt

masterpiece, best quality, 1girl, solo, long blue hair, white dress, underwater, bubbles, looking at viewer

Scene

masterpiece, best quality, 1boy, 1girl, rooftop, city lights, night, wind, standing, from behind, detailed background

Creative idea: flat graphic style

masterpiece, best quality, flat color, black outline, abstract background, 1girl, solo, profile, geometric shapes, limited palette

Tips

  • Over-tagging is the number one cause of blurry output on this model; remove tags until it sharpens.
  • Background tags (city at night, library, beach) are followed closely in v16.
  • Style
    LoRAs
    for
    Illustrious
    load cleanly at 0.6-0.8.
  • Flat color and bold outline prompts give a graphic poster look.

Felsökning

Blurry, soft image

Varför det händer
Too many quality and aesthetic tags or an overlong negative prompt.

Så löser du det
Cut the prompt down to the essentials and keep the negative to a few words.

Broken limbs

Varför det händer
Single-pass generation.

Så löser du det
Turn on hires fix at 1.5x; v16 and v17 were tuned so this pass repairs limbs.

AD
PirateDiffusion
Inget att installera - kör AI-modeller i molnet
PirateDiffusion finns bara på Telegram och är byggt för proffs: tusentals modeller, LoRA och arbetsflöden som styrs med chattkommandon, med obegränsad generering till ett fast pris.

Frågor

Which base is it?

Illustrious XL, so Illustrious and most NoobAI LoRAs work.

Does it understand natural language?

Partly. Tags remain the most reliable way to control it.

Länkar och källor

Modeller i den här guiden

Skriven av
Captain

Relaterade guider

Modeller
Playground v2 1024 Aesthetic: a second opinion to SDXL
Playground v2 is an SDXL-architecture model trained from scratch for aesthetics. Learn how it differs from SDXL, its CFG and step preferences, and the painterly and realistic prompts where it beats the original.
Av Captain
2026-10-02
Modeller
DreamShaper XL 1.0: versatile fantasy and realism on SDXL
Lykon's DreamShaper XL 1.0 brings the famous DreamShaper look to SDXL: fantasy, portraits, cinematic scenes and illustration in one model. Learn its settings, prompt style and creative uses.
Av Captain
2026-10-01
Modeller
Smooth Detail Booster: cleaner shading and more detail on Illustrious and Pony
DigitalPastel's Smooth Detail Booster LoRA adds detail and smooth shading to anime checkpoints without changing the art style. Learn the weight range, the prompts that help it, and how to pair it with quality tags.
Av Captain
2026-09-29
Modeller
Hyper-SD XL: one-to-eight-step SDXL with a flexible LoRA
Hyper-SD from ByteDance accelerates SDXL checkpoints to as few as one step. Learn which variant to pick, the CFG and sampler rules, and how to combine it with checkpoints and LoRAs without losing quality.
Av Captain
2026-10-02

Fler guider

Modeller
FLUX.1 Dev: how to prompt it, best settings and creative ideas
A practical guide to FLUX.1 Dev by Black Forest Labs: natural-language prompting, guidance and step settings, LoRA stacking, text rendering and prompt ideas that play to its strengths.
Av Captain
2026-09-30
Modeller
Stable Diffusion XL 1.0: prompts, settings and what it still does best
How to get the most out of the official SDXL 1.0 base model: native resolutions, CFG and sampler settings, the refiner, negative prompts and ideas for styles where SDXL still shines.
Av Captain
2026-10-01
Modeller
Stable Diffusion 1.5: the classic model, prompted properly
Stable Diffusion 1.5 is still worth knowing: the right resolution, CFG and sampler, how to use its huge library of LoRAs and embeddings, and creative prompt ideas suited to a 512-pixel model.
Av Captain
2026-10-02
Modeller
FLUX.2 Dev: prompting the newest Black Forest Labs model
FLUX.2 Dev brings larger prompts, better text, stronger realism and multi-reference editing. This guide covers the turbo workflow, guidance and resolution settings, prompt structure and ideas that use its new strengths.
Av Captain
2026-10-03
Modeller
Qwen-Image: long prompts, perfect text and bilingual posters
Qwen-Image is the model to reach for when the words in the picture matter. Learn how to write its long descriptive prompts, render English and Chinese text accurately, set CFG and steps, and explore layout-heavy creative ideas.
Av Captain
2026-09-29
Modeller
SDXL-Lightning: four-step generation on any SDXL checkpoint
ByteDance's SDXL-Lightning LoRA turns a 30-step SDXL render into a 4-step one. Learn the step counts, the CFG you must use, the sampler, and how to combine it with your favorite checkpoints and style LoRAs.
Av Captain
2026-09-30