nightmareai
/
disco-diffusion
Generate images using a variety of techniques - Powered by Discoart
Run nightmareai/disco-diffusion with an API
Input schema
Use RN50 model
- Default
- true
Seed (leave empty to use a random seed)
- Maximum
- 4294967295
Use RN101 model
Number of steps, higher numbers will give more refined output but will take longer
- Default
- 100
Width of the output image, higher numbers will take longer
- Default
- 1280
Use RN50x4 model
Use ViTB16 model
- Default
- true
Use ViTB32 model
- Default
- true
Use ViTB14 model
Height of the output image, higher numbers will take longer
- Default
- 768
Text Prompt
- Default
- "A beautiful painting of a singular lighthouse, shining its light across a tumultuous sea of blood by greg rutkowski and thomas kinkade, Trending on artstation."
Use RN50x16 model
Use RN50x64 model
Use RN50x101 model
TV Scale
Saturation Scale
Skip Augmentations
Use RN50_cc12m model
Use ViTL14_336 model
Initial image to start generation from
Initial Scale
- Default
- 1000
Skip Steps
- Default
- 10
Range Scale
- Default
- 150
Cut Batches
- Default
- 4
Steps between outputs, lower numbers may slow down generation.
- Default
- 20
Target image to generate towards, similarly to the text prompt
Target Scale
- Default
- 20000
Use RN101_yfcc15m model
Use RN50_yffcc15m model
Diffusion Model
- Default
- "512x512_diffusion_uncond_finetune_008100"
Use ViTB32_laion2b_e16 model
CLIP Guidance Scale
- Default
- 5000
Use secondary model
- Default
- true
Use RN50_quickgelu_cc12m model
Use ViTB16_laion400m_e31 model
Use ViTB16_laion400m_e32 model
Use ViTB32_laion400m_e31 model
Use ViTB32_laion400m_e32 model
Use RN50_quickgelu_yfcc15m model
Use RN101_quickgelu_yfcc15m model
Diffusion Sampling Mode
- Default
- "ddim"
Use ViTB32quickgelu_laion400m_e31 model
Use ViTB32quickgelu_laion400m_e32 model
Output schema
- Type
- uri[]