Nighttime Railway Portrait with SDXL Settings
From Wikiprompt, the free prompt encyclopedia
Nighttime Railway Portrait with SDXL Settings A complete Stable Diffusion prompt for a moody nighttime railway portrait, including negative prompt, model, sampler, and ControlNet/IP-Adapter settings.
Related media
Prompt ContentSave
🌐
A vertical nighttime photograph of a young woman standing outdoors next to metal stairs near a railway. She has light skin, brown hair pulled back into a low ponytail with loose strands framing her face and falling over her forehead, defined eyebrows, and a serious, pensive expression. She is wearing an oversized brown leather bomber jacket with visible zipper and ribbed cuffs, a black crew-neck t-shirt underneath, and a pleated olive green military-style mini skirt. She stands slightly turned, with her arms crossed over her chest, looking off to the left side of the frame, avoiding eye contact with the camera. The setting is an industrial area near train tracks: metal stairs with white railing on the right going up an embankment, metal catenary poles and railway electrification structures in the background, an electrical control box on the upper right, and a tall street light pole on the upper left. The steep slope behind her is covered in green grass and dirt. The night sky is dark grey-purple. Strong artificial light from a street lamp in the upper left corner creates a slight lens flare, casting hard direct light on her face and jacket. Shot from a slightly low angle, American shot framing. Photorealistic, high detail, cinematic lighting, moody urban night atmosphere, 8k, masterpiece, best quality, ultra-detailed, sharp focus.
Negative Prompt: cartoon, anime, illustration, painting, drawing, 3d render, cgi, deformed, distorted, bad anatomy, disfigured, poorly drawn face, mutation, mutated, extra limb, ugly, poorly drawn hands, missing limb, floating limbs, disconnected limbs, malformed hands, out of focus, extreme zoom, absurd, blurry, wet, watermark, signature, text, username, artist name, low quality, worst quality, normal quality, lowres, jpeg artifacts, dead, out of frame, cropped, mutation, mutilated, out of frame, extra fingers, too many fingers, cross-eyed, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, monochrome, grayscale, duplicate, morbid, mutilated, out of frame, extra toes, crossed eyes.
Model: sd_xl_base_1.0.safetensors or Juggernaut XL v9
Resolution: 832x1216 (9:16 aspect ratio)
Steps: 30-40
Sampler: DPM++ 2M Karras or Euler a
CFG Scale: 7-8
Clip Skip: 2 (for SD 1.5) / 1 (for SDXL)
Denoising Strength: 0.75 (if using img2img)
IP-Adapter Weight: 0.6-0.8
IP-Adapter Model: ip-adapter-plus_sd15.safetensors
Reference Image: original image of the model
ControlNet Unit 0:
Model: control_v11p_sd15_openpose
Preprocessor: openpose_full
Weight: 0.8-1.0
Guidance Start: 0.0
Guidance End: 1.0
ControlNet Unit 1:
Model: control_v11f1p_sd15_depth
Preprocessor: depth_midas
Weight: 0.4-0.6
InstantID Weight: 0.8
IdentityNet Scale: 0.5-0.8
Preserve Face Details: Enabled
add_detail.safetensors (Weight: 0.4-0.6)
epi_noiseoffset2.safetensors (Weight: 0.3-0.5)
cinematic_lighting.safetensors (Weight: 0.5)
Sign in to see the full prompt
Continue with:
By logging in, you agree to our Terms of Use and Privacy Policy
Usage
This prompt is designed for use with creative. Copy the prompt content above and paste it into your preferred AI tool.
For best results, you may customize the placeholders (indicated by square brackets or capital letters) with your specific requirements.
Talk
0 comments