sdxl-based/ays-text-to-image 🔢📝❓ → 🖼️

▶️ 9.5K runs 📅 Apr 2024 ⚙️ Cog 0.9.5 🔗 GitHub 📄 Paper ⚖️ License

sdxl text-to-image

About

Uses 'Align your steps' for faster higher quality images

Example Output

Prompt:

"a beautiful portrait photo of a cat"

Output

Performance Metrics

1.24s Prediction Time

1.25s Total Time

All Input Parameters

{
  "steps": 10,
  "width": 1156,
  "height": 768,
  "prompt": "a beautiful portrait photo of a cat",
  "checkpoint": "ProteusV0.4.safetensors",
  "num_outputs": 1,
  "sampler_name": "euler",
  "output_format": "webp",
  "guidance_scale": 7.5,
  "output_quality": 80,
  "negative_prompt": ""
}

Input Parameters

seed Type: integer
steps Type: integerDefault: 10Range: 10 - 100: Number of diffusion steps. (A minimum of 10 with AYS)
width Type: integerDefault: 1024
height Type: integerDefault: 1024
prompt Type: stringDefault: a photo of an astronaut riding a unicorn
checkpoint Default: albedobaseXL_v21.safetensors: The SDXL model to use for generation
num_outputs Type: integerDefault: 1Range: 1 - 10: Number of outputs
sampler_name Default: euler
output_format Default: webp: Format of the output images
guidance_scale Type: numberDefault: 7.5Range: 0 - 30: Scale for classifier-free guidance
output_quality Type: integerDefault: 80Range: 0 - 100: Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.
negative_prompt Type: stringDefault:: The negative prompt to guide image generation.

Output Schema

Output

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

Example Execution Logs

Random seed set to: 3470234499
Checking inputs
====================================
Checking weights
✅ ProteusV0.4.safetensors
====================================
Running workflow
got prompt
Executing node 21, title: SamplerCustom, class type: SamplerCustom
  0%|          | 0/10 [00:00<?, ?it/s]
 20%|██        | 2/10 [00:00<00:00, 16.72it/s]
 40%|████      | 4/10 [00:00<00:00, 13.17it/s]
 60%|██████    | 6/10 [00:00<00:00, 12.32it/s]
 80%|████████  | 8/10 [00:00<00:00, 11.94it/s]
100%|██████████| 10/10 [00:00<00:00, 11.76it/s]
100%|██████████| 10/10 [00:00<00:00, 12.21it/s]
Executing node 8, title: VAE Decode, class type: VAEDecode
Executing node 9, title: Save Image, class type: SaveImage
Prompt executed in 1.14 seconds
outputs:  {'9': {'images': [{'filename': 'ComfyUI_00001_.png', 'subfolder': '', 'type': 'output'}]}}
====================================
Contents of /tmp/outputs:
ComfyUI_00001_.png

Version Details

Version ID: a004c3ac8f62ac95a90b5a0c264beb47b66a6d1f8141b76fb27cd90e9a8bfe8e
Version Created: April 30, 2024

Run on Replicate →