Model
Trends
.ai
Model
Trends
.ai
the best open source ai models

CyberRealistic Pony v17: photographic portraits with Pony control

Topic: Models
By Captain
Published 2026-10-02
Share
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
DPM++ 2M SDE
Steps
30
Guidance
4
Seed
447612118
Size
832x1216
Clip skip
2
Prompt
masterpiece, best quality, ultra detailed, photorealistic, cinematic portrait, medium close up, eye level shot, 85mm lens, f1.8, shallow depth of field, 1woman, young ginger woman, delicate features, fair skin, serene expression, looking away, head slightly tilted upward, ((arm behind head)), natural makeup, soft blush, subtle lip sheen, ultra detailed skin texture, visible pores, cream lace blouse, intricate lace pattern, pearl stud earrings, long hair gently lifted by breeze, golden hour, warm amber backlight, soft rim lighting, glowing hair edges, volumetric light rays, light haze, high dynamic range, sunlit meadow, tall grass, creamy bokeh, pastel sky, soft drifting clouds, warm color harmony, cinematic composition, balanced exposure, sharp focus on subject, dreamy atmosphere, natural warmth
Negative prompt
(worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
Settings
Sampler
Cyberdelia Ralston (RK2)
Steps
24
Guidance
4.5
Seed
14533328
Size
2496x3648
Clip skip
2
Prompt
1girl, solo, blue eyes, white hair, messy hair, shoulder length hair, parted lips, blush, nose blush, parted lips, black eyeliner, makeup, close up, portrait, subject focus, looking at viewer, soft skin texture, natural skin detail, asymmetrical clothes, oversized clothes, loose clothes, collarbone, red background, cinematic lighting, dramatic lighting, soft shadows, high contrast, shallow depth of field, 85mm lens, photorealistic, ultra detailed, high resolution, film grain
Negative prompt
worst quality, bad quality, 3 fingers, bad hands, missing fingers
Settings
Sampler
Euler a
Steps
28
Guidance
8
Seed
3645456011
Size
1024x1024
Prompt
score_9, score_8_up, score_7_up, score_6_up BREAK source_anime BREAK rating_questionable BREAK chipmunk_girl, angel_costume, first_high_school_uniform, two-tone_background, 1girl, long_hair, slender, yellow_eyes, multicolored_eyes, red_eyes, curly_hair, multicolored_hair, orange_hair, purple_hair, grey_hair, dutch_angle BREAK hand_on_hip, looking_at_viewer BREAK sci-fi BREAK chiaroscuro, highly_detailed, eerie, landscape
Negative prompt
score_5_up, score_4_up, score_6, score_5, score_4, ugly, deformed, blurry, lowres, low_quality, bad_anatomy, mutated, long_neck, bad_proportions, 3d, cgi, render, low_detail, flat, simple, stock_photo, text, error, watermark, username, border, chromatic_aberration
Settings
Sampler
DPM++ 2M SDE
Steps
28
Guidance
4
Seed
3942105687
Size
832x1216
Clip skip
2
Prompt
1girl, solo, rapunzel \(disney\) ((absurdly long braided hair)) golden hair, thick braid, wide hips hair draped over shoulder floor length hair reaching the ground standing by tower window, looking out window wistful expression, soft gaze elegant purple dress, flowing fabric fantasy setting, stone tower interior soft magical illumination, gentle atmospheric light cinematic composition, soft glow, subtle depth of field fantasy realism, high quality, detailed, masterpiece
Negative prompt
(worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
Settings
Sampler
DPM++ 2M SDE
Steps
30
Guidance
4
Seed
2923687068
Size
832x1216
Clip skip
2
Prompt
red fox, animal, standing alert, full body, symmetrical composition, detailed fur texture, snowflakes on fur, snowy forest, pine trees, winter landscape, falling snow, soft sunlight, ambient twilight glow, cinematic lighting, shallow depth of field, magical bokeh, glowing particles, soft focus, photorealistic, ultra-detailed textures, realistic wildlife photography, natural lighting, peaceful atmosphere, magical realism, serene mood, high detail, masterpiece, best quality
Negative prompt
(worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
Settings
Sampler
DPM++ 2M SDE
Steps
30
Guidance
4
Seed
556810
Size
832x1216
Clip skip
2
Prompt
(masterpiece, best quality), high resolution, ultra detailed black and white vintage photography, medium format film aesthetic old abandoned car, centered in frame, rear view facing away from viewer, strict symmetrical composition once-proud silhouette now decaying, faded peeling paint revealing oxidized metal, deep rust blooms along chrome trim and wheel wells, rear windshield completely shattered, spiderweb cracks radiating from central impact point, missing glass shards, deflated tires sagging onto heavily rusted steel rims, thick dried mud and dust coating lower panels and bumpers deeply embedded in vast overgrown grassy field, tall dry unkempt grass reaching window height, gently swaying mid-ground dilapidated rural farmhouse, weathered gray wooden siding, broken windows, sagging partially collapsed roof, distant skeletal trees with gnarled branches, leaning utility poles silhouetted against sky uniform heavy overcast sky, flat diffused lighting, no harsh shadows, strong tonal contrast, pronounced film grain, detailed texture emphasis, deep depth of field, sharp focus throughout, melancholic desolate atmosphere, profound loneliness, time-worn stillness
Negative prompt
(worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
Settings
Sampler
Euler a
Steps
19
Guidance
7
Seed
1746728109
Size
832x1216
Clip skip
2
Weight used
1
Prompt
score_9, score_8_up, score_7_up, _ Masterpiece, best quality, high resolution, detailed, very detailed, intricate detailed, ___ cabbage, cjdl, made of green leaves, violet leaves, white veins, organic, ___ (bird:1.2),
Negative prompt
score_5, score_4, score_3, _ bad anatomy, bad hands, ugly, missing limbs, _ worst quality, low quality, watermark, broken, error, blurry, blurred, signature, deformed, lowres, bad anatomy, jpeg artifacts, _ nsfw, partial nudity, nude, naked, nudity, sexy attire, cleavage, nipples, breasts, navel, sexual, seductive _ simple background,
Settings
Sampler
DPM++ 2M SDE
Steps
28
Guidance
4
Seed
399772304
Size
832x1216
Clip skip
2
Prompt
masterpiece, best quality, ultra high resolution, photorealistic, fashion editorial portrait, 1woman, solo, tall woman, athletic build, curvy yet slender, elegant pose, sitting on designer couch, low angle shot, long slender legs, bare feet, delicate toes, narrow waist, crisp white shirt, slightly unbuttoned shirt, fitted black pants, black choker, silver ring choker, very long blonde hair, smooth hair, soft waves, bangs, striking blue eyes, full lips, pale skin, natural skin glow, subtle makeup, high fashion model face, confident expression, looking at viewer, vibrant futuristic loft apartment, sleek interior design, loft windows, bold architectural lines, layered materials, gradient background, warm vibrant colors, high contrast composition, soft focus background, pronounced depth of field, 85mm lens, cinematic lighting, rim lighting, soft shadows, sharp focus, ultra detailed, realistic skin texture, luxury aesthetic
Negative prompt
(worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark

Overview

CyberRealistic
Pony
v17 is a release of Cyberdelia's realistic Pony
checkpoint
that many people keep for its portrait rendering: natural skin, soft cinematic light and accurate lens behavior with 85mm-style prompts.
Settings follow the series: DPM++ SDE Karras / 2M Karras / Euler a, 30+ steps, 896x1152 or 832x1216,
CFG
5, Clip Skip 2, score chain first.

Reference

Name
Type
What it is
Prefix
score_9, score_8_up, score_7_up
Plus rating_safe.
Negative
text
score_6, score_5, score_4, (worst quality:1.2), (low quality:1.2), lowres, bad anatomy, bad hands, watermark
5
Steps /
sampler
30+, DPM++ SDE Karras
Resolution / Clip Skip
896x1152, 2
AD
No install required - run AI models on the cloud
BitVector Prism is the easiest way to start: pick a model, type a prompt and generate in a clean web app, with nothing to configure. BitVector is also available on Discord and on the web (SpyGlass).

Step by step

  1. Score chain, rating_safe, then photorealistic, cinematic portrait.
  2. Add framing and lens: medium close up, eye level, 85mm, f1.8, shallow depth of field.
  3. Describe the person plainly with age and features.
  4. CFG 5, 32 steps, DPM++ SDE Karras, Clip Skip 2.
  5. Hires 1.5x at 0.3.

Examples

Cinematic portrait

score_9, score_8_up, score_7_up, rating_safe, masterpiece, photorealistic, cinematic portrait, medium close up, eye level, 85mm, f1.8, shallow depth of field, a young ginger woman with delicate features and fair skin, serene expression, soft window light

Soft close-up

score_9, score_8_up, score_7_up, rating_safe, close up portrait of a woman with white shoulder-length messy hair, blue eyes, light blush, black eyeliner, soft skin texture, natural light

Creative idea: working portrait series

score_9, score_8_up, score_7_up, rating_safe, environmental portrait of a potter at her wheel, clay on hands, north light, 50mm

Tips

  • Lens language is followed closely; it is the easiest way to control the look.
  • Natural light descriptions (window, overcast) give its best skin.
  • The Positive embedding adds polish for close-ups.
  • Anime-ish prompts also work because of the Pony base; add source_anime if you want that.

Troubleshooting

Plastic skin

Why it happens
CFG above 6.

How to fix it
CFG 5 and natural light words.

Eyes misaligned

Why it happens
Small faces.

How to fix it
Closer framing or face inpaint.

AD
PirateDiffusion
No install required - run AI models on the cloud
PirateDiffusion is Telegram only and built for pros: thousands of models, LoRAs and workflows driven by chat commands, with unlimited generation on a fixed price plan.

Questions

v17 or v18?

v17 for soft portraits, v18 CoreShift for sharper textures.

Embeddings?

The CyberRealistic Positive and Negative Pony embeddings.

Links and sources

Models in this guide

Written by
Captain

Related guides

Models
Hassaku XL (Illustrious) v3.1: bright, distinct anime without Danbooru clutter
Ikena's Hassaku XL is a bright anime model with a clear house style. Learn why it avoids Danbooru metadata tags, its CFG and negative advice, and prompt ideas for vivid characters.
By Captain
2026-10-01
Models
CyberRealistic XL v10: super clean photorealism on SDXL
Cyberdelia's CyberRealistic XL is the SDXL checkpoint for clean, sharp photoreal images. Learn the recommended DPM++ 2M SDE Karras settings, the DMD2 fast path, upscaling advice and prompt ideas.
By Captain
2026-10-03
Models
Nova Anime XL (Illustrious): colorful anime, 2.5D and fantasy scenery
Crody's Nova Anime XL is a vivid Illustrious anime checkpoint. Learn its Euler a settings, quality tag strings, negative prompt, licence note and prompt ideas for colorful characters and landscapes.
By Captain
2026-09-29
Models
iLustMix v7 Cinematic: semi-realistic Illustrious with film lighting
GZees' iLustMix v7 Cinematic blends Illustrious anatomy with a cinematic, semi-realistic finish. Learn its Euler a / 25-step settings, quality tags and ideas for moody, movie-like frames.
By Captain
2026-09-29

More guides

Cloud services
Flat fee vs tokens: why unlimited plans like Graydient.ai are the best value for AI creators in 2026
A ranked cost comparison, with prices checked on 8 October 2026, of flat-fee unlimited plans like Graydient.ai against token and credit pricing from Midjourney, Leonardo, fal.ai, Replicate, Runway and Kling: what 1,000 images and 1,000 videos a month really cost on each, and why one fixed fee for unlimited images, video, audio, Grok LLM chat and web apps is the best value for creators who iterate.
By Captain
2026-10-08
Models
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.
By Captain
2026-09-30
Models
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.
By Captain
2026-10-01
Models
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.
By Captain
2026-10-02
Models
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.
By Captain
2026-10-03
Models
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.
By Captain
2026-09-29