fofr/latent-consistency-model 🔢🖼️📝✓❓ → 🖼️

▶️ 1.5M runs 📅 Oct 2023 ⚙️ Cog 0.8.6 🔗 GitHub ⚖️ License

controlnet image-to-image text-to-image

About

Super-fast, 0.6s per image. LCM with img2img, large batching and canny controlnet

Example Output

Prompt:

"Self-portrait oil painting, a beautiful cyborg with golden hair, 8k"

Output

Performance Metrics

0.85s Prediction Time

1.29s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/JlG0Efd2ubBp9yGnlOi7I9Se2rXnJSrPFogLf0YieKgjnWN6/download-6.png",
  "width": 768,
  "height": 768,
  "prompt": "Self-portrait oil painting, a beautiful cyborg with golden hair, 8k",
  "num_images": 1,
  "guidance_scale": 8,
  "archive_outputs": false,
  "prompt_strength": 0.45,
  "sizing_strategy": "width/height",
  "lcm_origin_steps": 50,
  "canny_low_threshold": 100,
  "num_inference_steps": 4,
  "canny_high_threshold": 200,
  "control_guidance_end": 1,
  "control_guidance_start": 0,
  "controlnet_conditioning_scale": 2
}

Input Parameters

seed Type: integer: Random seed. Leave blank to randomize the seed
image Type: string: Input image for img2img
width Type: integerDefault: 768: Width of output image. Lower if out of memory
height Type: integerDefault: 768: Height of output image. Lower if out of memory
prompt Type: stringDefault: Self-portrait oil painting, a beautiful cyborg with golden hair, 8k: For multiple prompts, enter each on a new line.
num_images Type: integerDefault: 1Range: 1 - 50: Number of images per prompt
control_image Type: string: Image for controlnet conditioning
guidance_scale Type: numberDefault: 8Range: 1 - 20: Scale for classifier-free guidance
archive_outputs Type: booleanDefault: false: Option to archive the output images
prompt_strength Type: numberDefault: 0.8Range: 0 - 1: Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
sizing_strategy Default: width/height: Decide how to resize images – use width/height, resize based on input image or control image
lcm_origin_steps Type: integerDefault: 50Range: 1 - ∞
canny_low_threshold Type: numberDefault: 100Range: 1 - 255: Canny low threshold
num_inference_steps Type: integerDefault: 8Range: 1 - 50: Number of denoising steps. Recommend 1 to 8 steps.
canny_high_threshold Type: numberDefault: 200Range: 1 - 255: Canny high threshold
control_guidance_end Type: numberDefault: 1Range: 0 - 1: Controlnet end
control_guidance_start Type: numberDefault: 0Range: 0 - 1: Controlnet start
disable_safety_checker Type: booleanDefault: false: Disable safety checker for generated images. This feature is only available through the API
controlnet_conditioning_scale Type: numberDefault: 2Range: 0.1 - 4: Controlnet conditioning scale

Output Schema

Output

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

Example Execution Logs

Using seed: 38477
Found 1 prompt:
- Self-portrait oil painting, a beautiful cyborg with golden hair, 8k
Making 1 image
Using given dimensions
img2img mode
  0%|          | 0/4 [00:00<?, ?it/s]
 50%|█████     | 2/4 [00:00<00:00, 17.91it/s]
100%|██████████| 4/4 [00:00<00:00, 17.82it/s]
100%|██████████| 4/4 [00:00<00:00, 17.81it/s]

Version Details

Version ID: 683d19dc312f7a9f0428b04429a9ccefd28dbf7785fef083ad5cf991b65f406f
Version Created: January 10, 2024

Run on Replicate →