fofr/sdxl-toy-story-people 🖼️🔢📝❓✓ → 🖼️

▶️ 3.0K runs 📅 Nov 2023 ⚙️ Cog 0.8.6
image-inpainting image-to-image text-to-image

About

SDXL fine-tuned on the people in Toy Story (1995)

Example Output

Prompt:

"An animated TOK Elon Musk person, 90s animation"

Output

Example output

Performance Metrics

11.14s Prediction Time
12.55s Total Time
All Input Parameters 
{
  "width": 1024,
  "height": 1024,
  "prompt": "An animated TOK Elon Musk person, 90s animation",
  "refine": "expert_ensemble_refiner",
  "scheduler": "K_EULER",
  "lora_scale": 0.6,
  "num_outputs": 1,
  "guidance_scale": 7.5,
  "apply_watermark": false,
  "high_noise_frac": 0.9,
  "negative_prompt": "broken, distorted, ugly, two people",
  "prompt_strength": 0.8,
  "num_inference_steps": 30
}
Input Parameters
mask Type: string
Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.
seed Type: integer
Random seed. Leave blank to randomize the seed
image Type: string
Input image for img2img or inpaint mode
width Type: integerDefault: 1024
Width of output image
height Type: integerDefault: 1024
Height of output image
prompt Type: stringDefault: An astronaut riding a rainbow unicorn
Input prompt
refine Default: no_refiner
Which refine style to use
scheduler Default: K_EULER
scheduler
lora_scale Type: numberDefault: 0.6Range: 0 - 1
LoRA additive scale. Only applicable on trained models.
num_outputs Type: integerDefault: 1Range: 1 - 4
Number of images to output.
refine_steps Type: integer
For base_image_refiner, the number of steps to refine, defaults to num_inference_steps
guidance_scale Type: numberDefault: 7.5Range: 1 - 50
Scale for classifier-free guidance
apply_watermark Type: booleanDefault: true
Applies a watermark to enable determining if an image is generated in downstream applications. If you have other provisions for generating or deploying images safely, you can use this to disable watermarking.
high_noise_frac Type: numberDefault: 0.8Range: 0 - 1
For expert_ensemble_refiner, the fraction of noise to use
negative_prompt Type: stringDefault:
Input Negative Prompt
prompt_strength Type: numberDefault: 0.8Range: 0 - 1
Prompt strength when using img2img / inpaint. 1.0 corresponds to full destruction of information in image
replicate_weights Type: string
Replicate LoRA weights to use. Leave blank to use the default weights.
num_inference_steps Type: integerDefault: 50Range: 1 - 500
Number of denoising steps
disable_safety_checker Type: booleanDefault: false
Disable safety checker for generated images. This feature is only available through the API. See https://replicate.com/docs/how-does-replicate-work#safety
Output Schema

Output

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

Example Execution Logs 
Using seed: 25576
Ensuring enough disk space...
Free disk space: 2068550201344
Downloading weights: https://replicate.delivery/pbxt/FoFPwBsrDfUBWq1rbROyhxqlncif4fghJN4I0AZgaFgzevSHB/trained_model.tar
b'Downloaded 186 MB bytes in 0.239s (779 MB/s)\nExtracted 186 MB in 0.058s (3.2 GB/s)\n'
Downloaded weights in 0.41710972785949707 seconds
Loading fine-tuned model
Does not have Unet. assume we are using LoRA
Loading Unet LoRA
Prompt: An animated <s0><s1> Elon Musk person, 90s animation
txt2img mode
  0%|          | 0/27 [00:00<?, ?it/s]
  4%|▎         | 1/27 [00:00<00:07,  3.69it/s]
  7%|▋         | 2/27 [00:00<00:06,  3.68it/s]
 11%|█         | 3/27 [00:00<00:06,  3.67it/s]
 15%|█▍        | 4/27 [00:01<00:06,  3.66it/s]
 19%|█▊        | 5/27 [00:01<00:06,  3.66it/s]
 22%|██▏       | 6/27 [00:01<00:05,  3.65it/s]
 26%|██▌       | 7/27 [00:01<00:05,  3.65it/s]
 30%|██▉       | 8/27 [00:02<00:05,  3.65it/s]
 33%|███▎      | 9/27 [00:02<00:04,  3.65it/s]
 37%|███▋      | 10/27 [00:02<00:04,  3.65it/s]
 41%|████      | 11/27 [00:03<00:04,  3.65it/s]
 44%|████▍     | 12/27 [00:03<00:04,  3.65it/s]
 48%|████▊     | 13/27 [00:03<00:03,  3.65it/s]
 52%|█████▏    | 14/27 [00:03<00:03,  3.65it/s]
 56%|█████▌    | 15/27 [00:04<00:03,  3.64it/s]
 59%|█████▉    | 16/27 [00:04<00:03,  3.64it/s]
 63%|██████▎   | 17/27 [00:04<00:02,  3.64it/s]
 67%|██████▋   | 18/27 [00:04<00:02,  3.64it/s]
 70%|███████   | 19/27 [00:05<00:02,  3.64it/s]
 74%|███████▍  | 20/27 [00:05<00:01,  3.64it/s]
 78%|███████▊  | 21/27 [00:05<00:01,  3.64it/s]
 81%|████████▏ | 22/27 [00:06<00:01,  3.64it/s]
 85%|████████▌ | 23/27 [00:06<00:01,  3.64it/s]
 89%|████████▉ | 24/27 [00:06<00:00,  3.64it/s]
 93%|█████████▎| 25/27 [00:06<00:00,  3.64it/s]
 96%|█████████▋| 26/27 [00:07<00:00,  3.64it/s]
100%|██████████| 27/27 [00:07<00:00,  3.64it/s]
100%|██████████| 27/27 [00:07<00:00,  3.65it/s]
  0%|          | 0/3 [00:00<?, ?it/s]
 33%|███▎      | 1/3 [00:00<00:00,  4.02it/s]
 67%|██████▋   | 2/3 [00:00<00:00,  4.14it/s]
100%|██████████| 3/3 [00:00<00:00,  4.18it/s]
100%|██████████| 3/3 [00:00<00:00,  4.16it/s]
Version Details
Version ID
e5603fd85f4cb9bffb1e49a1a1add5f3fcf3f2d5383bc84fd3ff7cc4fe5beb10
Version Created
November 3, 2023
Run on Replicate →