p-image-upscale

Prerequisites

Install and load these skills before generating (skip if already in context via @pruna):

Or install the full suite once: npx skills add PrunaAI/pruna-skills@pruna -y

Follow each skill's Before generating / craft sections — do not restate guide content here.

Agent habit

In the first reply, name `p-image-upscale` in backticks, confirm PRUNA_API_KEY, then ask for the source image + target megapixels (open intake → generation-diversity clarification intake if output size is unclear). Redirect new photo generation requests to p-image.

When NOT to use

Use a different skill instead:

HTTP (curl)

Upload source image

curl -X POST "https://api.pruna.ai/v1/files" \
  -H "apikey: ${PRUNA_API_KEY}" \
  -F "content=@/path/to/source.png"

Use urls.get as input.image.

Create (async — recommended)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-image-upscale' \
  -d '{
    "input": {
      "image": "https://api.pruna.ai/v1/files/FILE_ID",
      "target": 8,
      "enhance_details": true,
      "output_format": "png"
    }
  }'

Poll and download: follow pruna-api.

Create (sync — quick test only)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-image-upscale' \
  -H 'Try-Sync: true' \
  -d '{
    "input": {
      "image": "https://api.pruna.ai/v1/files/FILE_ID",
      "target": 4,
      "enhance_details": true,
      "enhance_realism": false,
      "output_format": "png"
    }
  }'

Before generating

1. Complete Prerequisites guide reading order.
2. Confirm target MP (1–128), enhance_details / enhance_realism, and output_format with the user.
3. Pruna note: defaults — enhance_details: true, enhance_realism: false. Use enhance_realism: true only on already-photoreal sources; it can add waxy artifacts on synthetic edits. Source must already pass the slop gate.

Print pipeline: p-image hero → optional p-image-edit → p-image-upscale on the approved plate (upscale after edits). Typical large-crop target: 8 megapixels — confirm target with user.

Required input

• image (string URL)

Common optional fields

• target: integer megapixels 1–128 (default 4). Confirm current limits on p-image-upscale model docs.
• output_format: jpg, png, webp
• output_quality: 0–100 (not used for PNG)
• enhance_details, enhance_realism (booleans; realism can drift more from source)
• disable_safety_checker

Typical next steps

Common follow-ons after this skill: