Try this model on one of your photos to see its result here.

Classification model

Names what a photo shows, with the five most likely of 1,000 labels and their confidence. The model behind /classify.

Tasks: Classification
Outputs: top 5 labels, confidence scores
Classes (1,000): The 1,000 ImageNet classes. Labels in results are always in English.

Specifications

Training data: ImageNet-1k (ILSVRC 2012)
Accuracy: Classification
Top-1 accuracy
88.0%
ImageNet-1k validation set, all 50,000 photos, measured by Comexp on the model as served · measured Nov 10, 2026
Image requirements: JPG, PNG or WebP, up to 10 MB. One main subject per photo works best.
Video requirements: Photos only.
Input / output: In: one image, multipart/form-data field file. Out: JSON predictions, five labels with confidence.
Licence: Use through the Comexp API under our Terms of service.
Version: Not stated
Hardware: Runs in the Comexp cloud: nothing to install, any device that can send an HTTP request. On your own servers or devices: on request.
API: curl -X POST https://ml.comexp.net/api/classify \ -H "Authorization: ApiKey <YOUR_API_KEY>" \ -F "file=@image.jpg"
Each call costs one detection from your plan, as on /detect. Your API key is in your account settings.