image-upscale
Image Upscale
Upscales an image using AI super-resolution models, increasing resolution while generating realistic detail.
Command
npx agent-media@latest image upscale --in <path> [options]
Inputs
| Option | Required | Description |
|---|---|---|
--in |
Yes | Input file path or URL |
--scale |
No | Scale factor (2 or 4, default: 2) |
--out |
No | Output path, filename or directory (default: ./) |
--provider |
No | Provider to use (local, fal, replicate) |
--model |
No | Model to use (overrides provider default) |
Output
Returns a JSON object with the upscaled image path:
{
"ok": true,
"media_type": "image",
"action": "upscale",
"provider": "local",
"output_path": "upscaled_123_abc.png",
"mime": "image/png",
"bytes": 456789
}
Examples
Upscale an image (default 2x):
npx agent-media@latest image upscale --in photo.jpg
Upscale 4x with specific provider:
npx agent-media@latest image upscale --in photo.jpg --scale 4 --provider fal
Upscale with custom output:
npx agent-media@latest image upscale --in photo.jpg --out ./upscaled
Providers
local
Runs locally on CPU using Transformers.js, no API key required.
- Uses
Xenova/swin2SR-compressed-sr-x4-48model (~1.3MB) - Always outputs 4x upscale regardless of
--scale(model architecture limitation) - Models downloaded on first use
- You may see a
mutex lock failederror — ignore it, the output is correct if"ok": true
npx agent-media@latest image upscale --in photo.jpg --provider local
fal
- Requires
FAL_API_KEY - Uses
fal-ai/esrgan(Real-ESRGAN) model - Supports 2x and 4x scale
replicate
- Requires
REPLICATE_API_TOKEN - Uses
nightmareai/real-esrganmodel - Supports 2-10x scale
More from agntswrm/agent-media
audio-transcribe
Transcribes audio to text with timestamps and optional speaker identification. Use when you need to convert speech to text, create subtitles, transcribe meetings, or process voice recordings.
131image-remove-background
Removes the background from an image, leaving the foreground subject with transparency. Use when you need to isolate subjects, create cutouts, or prepare images for compositing.
129image-crop
Crops an image to specified dimensions around a focal point. Use when you need to extract a portion of an image, create thumbnails with custom positioning, or prepare images for specific aspect ratios.
83image-generate
Generates an image from a text prompt using AI models. Use when you need to create images from descriptions, generate artwork, or produce visual content from text.
78video-generate
Generates video from text prompts or animates static images. Use when you need to create videos from descriptions, animate images, or produce video content using AI.
68image-edit
Edits an existing image using a text prompt. Use when you need to modify, enhance, or transform an image based on text instructions.
62