AIRealNet

Model overview

AIRealNet is published by Modotte as an image-classification model. Captured configuration identifies Swinv2ForImageClassification with model type swinv2, and Safetensors metadata reports 195,206,006 parameters. According to the model card, it is a binary classifier for AI-generated versus real human images built on a SwinV2 Tiny backbone.

Recorded capabilities

Use cases in the source record

• Screening conventional fully generated images versus real photographs with the documented image-classification pipeline and Class 0 versus Class 1 outputs.
• According to the model card, use should avoid reliance on subtle nano-scale edit detection, where the publisher reports limited capability.

Limitations and unknowns

• According to the model card, the model struggles with subtle nano-scale edits, and very high controlled-dataset accuracy may be lower in real-world image domains.
• According to the model card, reported performance metrics come from epoch 2 as a stable post-fine-tuning illustration.
• No independent evaluation results were extracted from this record.
• Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Modotte/AIRealNet

Captured: Unknown. Processed: 2026-09-07T19:35:41.902584+00:00.

Modotte GitHub Repository Live Demo Overview In an era of rapidly advancing AI-generated imagery, deepfakes, and synthetic media, the need for reliable detection tools has never been higher. AIRealNet is a binary image classifier explicitly designed to distinguish AI-generated images from real human photographs . This model is optimized to detect conventional AI-generated content while adhering to strict privacy standards—avoiding personal or sensitive images. Class 0: AI-generated image Class 1: Real human image By leveraging the robust SwinV2 Tiny architecture as its backbone, AIRealNet achieves a high degree of accuracy while rem…