ReplicateImage to imageComing soon

retro-diffusion/rd-fast

Name: retro-diffusion/rd-fast
Brand: Replicate

retro-diffusion/rd-fast is an image to image tool from Replicate that runs in the background and returns images. Coming soon: listed with its price, not yet open to runs.

This tool is listed but not yet available to run.

Priced by usage

Pricing

$0.0546 per image. Billed on num_images, default 1. The final charge comes from usage.

Limits

Runs: In the background
Returns: image
Longest queue wait: 3600s
Run deadline: 1800s
Largest inline input: 1024 KB
Largest input file: 500 MB
Largest output: 4768 MB
Outputs per run: 16

Input

Field	Kind	Default	Description
`prompt` required	text	—	Text prompt for generation
`seed`	number (whole)	—	Random seed. Set for reproducible generation
`style`	JSON	default	Style to apply to the generated image default: Simple clean pixel art, with Anime illustration influences simple: Simple shading with minimalist shapes and designs detailed: Pixel art with lots of shading and details retro: A classic arcade game aesthetic inspired by early PC games game_asset: Distinct assets set on a simple background portrait: Character portrait focused images with high detail texture: Flat game textures like stones, bricks, or wood ui: User interface boxes and buttons item_sheet: Sheets of objects placed on a simple background character_turnaround: Character sprites viewed from different angles 1_bit: Two color black and white only images low_res: General low resolution pixel art images mc_item: Minecraft-styled items with automatic transparency mc_texture: Minecraft-styled flat textures, like grass, stones, or wood no_style: Pixel art with no style influence applied
`width`	number (whole) 16 to 384	256	Image width
`height`	number (whole) 16 to 384	256	Image height
`tile_x`	true or false	false	Enable seamless tiling on X axis
`tile_y`	true or false	false	Enable seamless tiling on Y axis
`strength`	number 0 to 1	0.8	Prompt strength in image-to-image generation. Higher values give closer adherence to the prompt, and lower values use more information from the input image. Only used when input_image is provided.
`remove_bg`	true or false	false	Remove background to create transparent images
`num_images`	number (whole) 1 to 10	1	Number of images to generate
`input_image`	text	—	Input image for image-to-image generation or reference. Will be converted to RGB without transparency.
`input_palette`	text	—	Reference palette image to guide color choices
`bypass_prompt_expansion`	true or false	false	Disable automatic prompt expansion

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "replicate",
    "model": "retro-diffusion/rd-fast",
    "input": {
      "prompt": "<prompt>"
    }
  }'