Create

Generate Similar

Generation Mode Choose how the image should start.

Prompt

{
  "prompt": "ultra photorealistic real camera portrait photo of a beautiful young Chinese woman idol, oval face, flawless smooth skin, long wet black hair, serene expression eyes gently closed, wearing flowing ethereal white lace dress, leaning gracefully against a brand new white lacquered grand piano partially submerged in shallow shimmering dark water, golden sunlight rays from above, delicate butterflies with translucent blue-pink wings fluttering around, floating pearl oysters glowing softly with pearl inside, cinematic golden hour lighting, warm amber tones, melancholic poetic atmosphere, high detail skin texture pores visible, natural wet hair strands, shallow depth of field, sharp focus on face hair and text elements",
  "negative_prompt": "cartoon, anime, illustration, painting, oil paint, watercolor, 3D render, deformed, blurry, low resolution, plastic skin, doll face, extra limbs, mutated hands, bad anatomy, watermark overlap, text distortion, AI generated look, artifacts, noise, grain, surreal glow, heavy fantasy, low contrast text, faint text, unreadable text, missing PF, blurred PF, weak PF, overexposed, underexposed",
  "style": "photorealistic, ultra realistic photography, pure real camera shot, no AI artifacts, lifelike, cinematic color grading",
  "camera": "Canon EOS R5, 85mm prime lens, f/1.4, professional portrait photography, 8K HDR, ultra detailed, meticulous realism",
  "lighting": "golden hour autumn sunlight, warm golden rays, soft cinematic lighting, dramatic but natural contrast",
  "composition": "vertical portrait, subject centered, shallow depth of field, bokeh background",
  "text_elements": {
    "highest_priority": {
      "text": "PF",
      "description": "massive bold thick black gothic font, 3D立体感, extremely prominent clear dominant in composition, behind the character, oversized, maximum legibility and boldness, sharp contrast, no blur, most important element in the entire image, weight:1.8-2.0"
    },
    "watermarks": [
      {
        "location": "top left corner",
        "text": "A",
        "style": "ultra small gothic handwritten crystal embossed font, 80% transparency, subtle press imprint"
      },
      {
        "location": "top right corner",
        "text": "I",
        "style": "ultra small gothic handwritten crystal embossed font, 80% transparency, subtle press imprint"
      },
      {
        "location": "top center",
        "text": "94vanAI",
        "style": "ultra small gothic handwritten crystal embossed font, crystal texture, subtle"
      },
      {
        "location": "bottom center",
        "text": "94vanAI",
        "style": "ultra small handwritten font, subtle"
      },
      {
        "location": "side panel of white grand piano body",
        "text": "94van",
        "style": "metallic embossed chrome engraved elegant craft font, realistic metallic shine, professional plate"
      }
    ]
  },
  "aspect_ratio": "16:9",
  "generation_tips": {
    "seed": "use fixed seed for consistency if supported",
    "steps": 40,
    "cfg_scale": 7.5,
    "sampler": "DPM++ 2M Karras or Euler a",
    "aspect_ratio": "3:4 or 4:5 (portrait)",
    "priority_order": [
      "1. Bold PF text legibility and size (absolute highest priority)",
      "2. All watermarks perfectly integrated and realistic",
      "3. Photorealistic human skin/hair quality",
      "4. Overall mood and lighting",
      "5. Butterflies and pearl oysters as secondary details"
    ]
  },
  "additional_stabilization": {
    "use_controlnet": "if available, use reference image + canny/pose for consistency",
    "denoising_strength": 0.35,
    "clip_skip": 1,
    "emphasis": "PF text:1.9, photorealistic:1.3, watermarks:1.2"
  }
}

3719 / 10000

Based on: Ethereal Harmony: A Butterfly Serenade

Model

Select a model to continue.

Each model exposes different quality and editing capabilities.

Size & Aspect Ratio

Pick the output framing.

Available options change with the selected model.

Select a model to see credit cost. Reference mode adds 1 credit on supported models.

Generate Similar

Ready when you are

Generating your image

Image ready

We could not finish this image

Generating your image

Image ready

We could not finish this image

Send people to the full prompt and image