fermatresearch/magic-image-refiner 🔢🖼️📝❓✓ → 🖼️

▶️ 945.9K runs 📅 Jan 2024 ⚙️ Cog 0.8.6 🔗 GitHub

image-inpainting image-to-image image-upscaling

About

A better alternative to SDXL refiners, providing a lot of quality and detail. Can also be used for inpainting or upscaling.

Example Output

Prompt:

"UHD 4k vogue, a woman resting in a magic pool, face above the surface of the water, red freckles"

Output

Performance Metrics

3.43s Prediction Time

3.45s Total Time

All Input Parameters

{
  "hdr": 0,
  "image": "https://replicate.delivery/pbxt/KA9yP9n3ZX5A5mkoPz3gsPzKTH1NA7LqVkQRTg7Sov46lOfo/0_1.webp",
  "steps": 20,
  "prompt": "UHD 4k vogue, a woman resting in a magic pool, face above the surface of the water, red freckles",
  "scheduler": "DDIM",
  "creativity": 0.25,
  "guess_mode": false,
  "resolution": "original",
  "resemblance": 0.75,
  "guidance_scale": 7,
  "negative_prompt": "teeth, tooth, open mouth, longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, mutant"
}

Input Parameters

hdr Type: numberDefault: 0Range: 0 - 1: HDR improvement over the original image
mask Type: string: When provided, refines some section of the image. Must be the same size as the image
seed Type: integer: Seed
image Type: string: Image to refine
steps Type: integerDefault: 20: Steps
prompt Type: string: Prompt for the model
scheduler Default: DDIM: Choose a scheduler.
creativity Type: numberDefault: 0.25Range: 0 - 1: Denoising strength. 1 means total destruction of the original image
guess_mode Type: booleanDefault: false: In this mode, the ControlNet encoder will try best to recognize the content of the input image even if you remove all prompts. The `guidance_scale` between 3.0 and 5.0 is recommended.
resolution Default: original: Image resolution
resemblance Type: numberDefault: 0.75Range: 0 - 1: Conditioning scale for controlnet
guidance_scale Type: numberDefault: 7Range: 0.1 - 30: Scale for classifier-free guidance
negative_prompt Type: stringDefault: teeth, tooth, open mouth, longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, mutant: Negative prompt

Output Schema

Output

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

Example Execution Logs

Using seed: 17908
  0%|          | 0/5 [00:00<?, ?it/s]
 20%|██        | 1/5 [00:00<00:01,  3.82it/s]
 40%|████      | 2/5 [00:00<00:00,  3.81it/s]
 60%|██████    | 3/5 [00:00<00:00,  3.81it/s]
 80%|████████  | 4/5 [00:01<00:00,  3.81it/s]
100%|██████████| 5/5 [00:01<00:00,  3.81it/s]
100%|██████████| 5/5 [00:01<00:00,  3.81it/s]

Version Details

Version ID: 507ddf6f977a7e30e46c0daefd30de7d563c72322f9e4cf7cbac52ef0f667b13
Version Created: January 31, 2024

Run on Replicate →