Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · Nahrawy

AIorNot

AIorNot is a Nahrawy Swin image classifier for real versus AI-generated images. According to the model card, it finetunes swin-tiny-patch4-window7-224.

Publisher
Nahrawy
Task
image-classification
Model type
swin
License
apache-2.0
Library
transformers
Publication status
Accepted · not indexed

Model overview

AIorNot is published by Nahrawy as a swin-based image-classification model. The captured configuration identifies SwinForImageClassification. According to the model card, it uses swin-tiny-patch4-window7-224 finetuned on the aiornot dataset to classify real and AI-generated images.

Recorded capabilities

Swin-tiny classification

Captured config identifies SwinForImageClassification, and the card describes swin-tiny-patch4-window7-224 finetuned on the aiornot dataset.

Real-versus-AI labels

According to the model card, the model classifies real and AI-generated images with Real and AI labels.

Documented Transformers loading

The card documents loading with AutoFeatureExtractor and AutoModelForImageClassification from Nahrawy/AIorNot.

Use cases in the source record

  • Real-versus-AI image-screening experiments using the card's documented Real and AI labels.

Limitations and unknowns

  • No parameter count was extracted from this record.
  • No evaluation results were extracted from this record.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Nahrawy/AIorNot

Captured: Unknown. Processed: 2026-09-07T19:34:34.612290+00:00.

Classification model used to classify real images and AI generated images. The model used is swin-tiny-patch4-window7-224 finetued on aiornot dataset. To use the model import torch from transformers import AutoFeatureExtractor, AutoModelForImageClassification labels = ["Real", "AI"] feature_extractor = AutoFeatureExtractor.from_pretrained("Nahrawy/AIorNot") model = AutoModelForImageClassification.from_pretrained("Nahrawy/AIorNot") input = feature_extractor(image, return_tensors="pt") with torch.no_grad(): outputs = model(**input) logits = outputs.logits prediction = logits.argmax(-1).item() label = labels[prediction]

F001F002F003F004F005F006F009F010F011F012F013