jichengdu/flux-pulid 🔢📝❓🖼️ → 🖼️

▶️ 337 runs 📅 Mar 2025 ⚙️ Cog 0.13.7 🔗 GitHub ⚖️ License

image-consistent-character-generation image-to-image

About

PuLID-FLUX-v0.9.0

Example Output

Prompt:

"portrait, color, cinematic"

Output

Performance Metrics

12.10s Prediction Time

234.42s Total Time

All Input Parameters

{
  "width": 896,
  "height": 1152,
  "prompt": "portrait, color, cinematic",
  "true_cfg": 1,
  "id_weight": 1,
  "num_steps": 20,
  "start_step": 0,
  "num_outputs": 1,
  "output_format": "webp",
  "guidance_scale": 4,
  "output_quality": 80,
  "main_face_image": "https://replicate.delivery/pbxt/Me8PvHwdoSXGlm8jtoxY2yHT254b6dV9EzWMLp29RjlGqT5E/43.jpg",
  "negative_prompt": "bad quality, worst quality, text, signature, watermark, extra limbs, low resolution, partially rendered objects, deformed or partially rendered eyes, deformed, deformed eyeballs, cross-eyed, blurry",
  "max_sequence_length": 128
}

Input Parameters

seed Type: integer: Set a random seed for generation (leave blank or -1 for random)
width Type: integerDefault: 896Range: 256 - 1536: Set the width of the generated image (256-1536 pixels)
height Type: integerDefault: 1152Range: 256 - 1536: Set the height of the generated image (256-1536 pixels)
prompt Type: stringDefault: portrait, color, cinematic: Enter a text prompt to guide image generation
true_cfg Type: numberDefault: 1Range: 1 - 10: Set the Classifier-Free Guidance (CFG) scale. 1.0 uses standard CFG, while values >1.0 enable True CFG for more precise control over generation. Higher values increase adherence to the prompt at the cost of image quality.
id_weight Type: numberDefault: 1Range: 0 - 3: Set the weight of the ID image influence (0.0-3.0)
num_steps Type: integerDefault: 20Range: 1 - 20: Set the number of denoising steps (1-20)
start_step Type: integerDefault: 0Range: 0 - 10: Set the timestep to start inserting ID (0-4 recommended, 0 for highest fidelity, 4 for more editability)
num_outputs Type: integerDefault: 1Range: 1 - 4: Set the number of images to generate (1-4)
output_format Default: webp: Choose the format of the output image
guidance_scale Type: numberDefault: 4Range: 1 - 10: Set the guidance scale for text prompt influence (1.0-10.0)
output_quality Type: integerDefault: 80Range: 1 - 100: Set the quality of the output image for jpg and webp (1-100)
main_face_image (required) Type: string: Upload an ID image for face generation
negative_prompt Type: stringDefault: bad quality, worst quality, text, signature, watermark, extra limbs, low resolution, partially rendered objects, deformed or partially rendered eyes, deformed, deformed eyeballs, cross-eyed, blurry: Enter a negative prompt to specify what to avoid in the image
max_sequence_length Type: integerDefault: 128Range: 128 - 512: Set the max sequence length for prompt (T5), smaller is faster (128-512)

Output Schema

Output

Type: array • Items Type: string • Items Format: uri

Example Execution Logs

Using seeds: [2602907503]
Generating 'portrait, color, cinematic' with seeds [2602907503]
Denoising time: 8.68 seconds
Decoding time: 0.37 seconds
Total generate_image time: 11.73 seconds
Image 1 generated with seed: 2602907503
Total prediction time: 11.89 seconds

Version Details

Version ID: ca0b96aa9f0d0a0a7962a727377806215df25a517865e5f6949f24a26c24d2a6
Version Created: March 12, 2025

Run on Replicate →