Model
Trends
.ai
Model
Trends
.ai
les meilleurs modèles d'IA open source

CyberRealistic Pony v13: the settings and prompts that make it shine

Sujet : Modèles
Par Captain
Publié le 2026-10-01
Version 13 of CyberRealistic Pony is a favorite for cinematic environments and natural portraits. Learn Cyberdelia's recommended settings, the score chain, and ideas for atmospheric scenes.
Cette page n'est pas encore traduite ; elle est affichée en anglais.

Aperçu

CyberRealistic
Pony
v13 is an earlier release of Cyberdelia's realistic Pony
checkpoint
, still popular for its balanced color and strong environment rendering. It follows the same recipe as the newer versions: Pony's score tags for quality control, realistic textures and moody cinematic light.
Settings are identical across the series: DPM++ SDE Karras or 2M Karras, 30+ steps, 896x1152 or 832x1216,
CFG
5, Clip Skip 2.

Référence

Nom
Type
Rôle
Prefix
score_9, score_8_up, score_7_up
First in the prompt; add rating_safe.
Negative
text
score_6, score_5, score_4, (worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
CFG
5
Author's value.
Steps / sampler
30+, DPM++ SDE Karras
Or Euler a.
Resolution / Clip Skip
832x1216, Clip Skip 2
Use landscape buckets (1216x832) for environments.
AD
Aucune installation requise – exécutez les modèles IA dans le cloud
BitVector Prism est la façon la plus simple de commencer : choisissez un modèle, tapez un prompt et générez dans une application web épurée, sans rien configurer. BitVector est aussi disponible sur Discord et sur le web (SpyGlass).

Pas à pas

  1. Score chain and rating_safe first.
  2. Describe the environment in layers: foreground, midground, background, then atmosphere.
  3. Use 1216x832 for wide scenes.
  4. CFG 5, 32 steps, DPM++ SDE Karras, Clip Skip 2.
  5. Hires fix 1.5x at 0.3 denoise.

Exemples

Environment

score_9, score_8_up, score_7_up, rating_safe, misty lane in a medieval village at night, dark tiled roofs, light walls, deciduous trees, distant pagoda, rocky mountains in fog, cinematic

Steampunk city

score_9, score_8_up, score_7_up, rating_safe, cliffside steampunk city, timber-framed buildings, gears and pipes, staircases, a bridge, two people on a boat, ultra detailed, atmospheric depth

Creative idea: weather portrait

score_9, score_8_up, score_7_up, rating_safe, photo of a lighthouse keeper in oilskins during a storm, spray, hard side light, 35mm

Astuces

  • v13 renders fog, mist and wet surfaces convincingly; ask for them.
  • Keep the score chain even for landscapes; it still controls quality.
  • For people, describe age and clothing plainly after the chain.
  • Pony composition tags (wide shot, from below) work.

Dépannage

Flat environments

Pourquoi cela arrive
No atmosphere words.

Comment le corriger
Add mist, volumetric light, depth haze and a named light source.

Soft faces in wide shots

Pourquoi cela arrive
Small faces at 1 MP.

Comment le corriger
Inpaint the face at full resolution or move to a closer framing.

AD
PirateDiffusion
Aucune installation requise – exécutez les modèles IA dans le cloud
PirateDiffusion est réservé à Telegram et conçu pour les pros : des milliers de modèles, LoRA et workflows pilotés par commandes de chat, avec génération illimitée à prix fixe.

Questions

Why use v13 instead of v18?

Preference. v13 has a slightly softer, more filmic grade that some people like for environments.

Same embeddings?

Yes, the CyberRealistic Positive and Negative Pony embeddings work on all versions.

Liens et sources

Modèles de ce guide

Écrit par
Captain

Guides associés

Modèles
Dramatic Lighting Slider: strong shadows and glowing highlights on Pony and Illustrious
klaabu's Dramatic Lighting Slider changes only the lighting, not the subject. Learn its weight ranges for anime and realistic checkpoints, the lighting words that pair with it and creative uses.
Par Captain
2026-10-03
Modèles
CyberRealistic Pony v18 CoreShift: realistic Pony prompting
Cyberdelia's CyberRealistic Pony mixes Pony's control tags with photographic realism. Learn the score chain, the exact settings the author recommends, negative prompt, the companion embeddings and creative ideas.
Par Captain
2026-10-02
Modèles
Prefect Illustrious XL v7: polished anime with soft lighting
Goofy_Ai's Prefect Illustrious XL v7 is tuned for clean, soft-lit anime illustration. Learn the author's settings (Clip Skip 1, Euler a, CFG 5-6), prompt template and creative ideas.
Par Captain
2026-10-02
Modèles
One Obsession v19 Atypical: beginner-friendly anime with hidden artist strings
maxfeifei8's One Obsession is an Illustrious/NoobAI anime checkpoint that makes good pictures without LoRAs or complex tags. Learn the clean illustration template, settings and how to discover the artist tags hidden in its examples.
Par Captain
2026-10-03

Autres guides

Modèles
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.
Par Captain
2026-09-30
Modèles
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.
Par Captain
2026-10-01
Modèles
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.
Par Captain
2026-10-02
Modèles
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.
Par Captain
2026-10-03
Modèles
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.
Par Captain
2026-09-29
Modèles
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.
Par Captain
2026-09-30