Portrait detail| Specification | Credits | Price (USD) |
|---|---|---|
| Varies by configuration | 0.6 – 14.4per generation | $0.006 – $0.144per generation |
Prices follow the model configuration. Reference images are not billed unless the model says so.
P-Image-Upscale
P-Image-Upscale is a production-focused super-resolution model for turning small or compressed images into cleaner, larger assets.
What it is for
Use it when you need a larger, cleaner image for product assets, creative review, social publishing, or restoration workflows.
Examples
View more examples ↗
Portrait detailSupported PrunaAI P-Image-Upscale Controls
| Control | Type | Default | Notes | |
|---|---|---|---|---|
| 01 | image_url | image | — | Input image to upscale. |
| 02 | upscale_mode | string | target | Upscale mode. 'target' scales to a fixed megapixel resolution. 'factor' multiplies each side by the given factor. Default: "target" |
| 03 | target | integer1 – 128 | 4 | Target resolution in megapixels (used when upscale_mode is 'target'). Default: 4 |
| 04 | factor | number1 – 8 | 2 | Scaling factor applied to each side of the image (used when upscale_mode is 'factor'). Output capped at 128 MP. Default: 2 |
| 05 | enhance_details | boolean | false | Enhance fine textures and small details. May increase contrast and introduce minor deviations from the original image. Default: false |
| 06 | enhance_realism | boolean | false | Improve realism. May deviate more from the original image. Recommended for AI-generated images. Default: false |
| 07 | output_format | string | jpg | Format of the output images Default: "jpg" |
| 08 | output_quality | integer0 – 100 | 80 | Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs Default: 80 |
Use PrunaAI P-Image-Upscale through the AIReiter API
API ↗prunaai_p_image_upscale
POST to /api/openapi/submit with the model, a unique out_task_id, and parameters.
POST to /api/openapi/query with the same out_task_id until the task completes.
Save outputs in your own storage rather than hot-linking them.
