Automatic image labels · AI beta
Automatic image labels · AI beta — 20 photos / English label candidates / JSON / CSV
Up to 20 images · 100 MiB total. Downloads a roughly 19 MB EfficientNet-Lite0 model and the shared engine. Save five candidates from 1,000 fixed English ImageNet labels as JSON / CSV. Predictions may be wrong; this does not generate free-form descriptions.
- Select up to 20 photos
- Review English label candidates
- Save labels as JSON / CSV
Usage example
- Example input
cat.jpg + car.jpg- Expected result
Five English label candidates and model scores per photo [review]
Examples are illustrative. Actual results depend on the source file and settings.
Examples and limitations
20 photos / English label candidates / JSON / CSV
Input and output
- Supported input
image/png · image/jpeg · image/webp · .png · .jpg · .jpeg · .webp- Saved output
JSON · CSV
Usage example
- Example input
cat.jpg + car.jpg- Expected result
Five English label candidates and model scores per photo [review]
Examples are illustrative. Actual results depend on the source file and settings.
Limits and checks
- Up to 20 images · 100 MiB total. Downloads a roughly 19 MB EfficientNet-Lite0 model and the shared engine. Save five candidates from 1,000 fixed English ImageNet labels as JSON / CSV. Predictions may be wrong; this does not generate free-form descriptions.
- PNG / JPG / WebP · 30 MiB / 12 MP per file. Originals stay unchanged; processing runs on your device.