AirasInnovations/Satyadrishti-V1-LARGE

Satyadrishti-V1 Large

🔭 Model Overview

Satyadrishti-V1 is an open-source image classification model by AIRAS INC, fine-tuned for deepfake detection. It is designed to classify images, determining if they are "Real" or "Fake". This model is part of the Satyadrishti series and leverages Google's google/siglip-so400m-patch14-384 image classification architecture.

This specific variant, Satyadrishti-V1 Large, has been fine-tuned for robust performance with its 840 million parameters.

📦 General Information

Metadata	Value
Model Name	Satyadrishti-V1 Large
Model ID	AirasInnovations/Satyadrishti-V1-LARGE
License	Apache-2.0
Language	English
Library	Transformers
Base Model	google/siglip-so400m-patch14-384
Pipeline Tag	image-classification
Tags	image-classification, deepfakedetection, image-authenticity

🧠 Model Architecture

Architecture: Fine-tuned SigLIP Vision Transformer
Task: Binary Image Classification (Real vs Fake)
Classes:
- Class 0: Fake (Deepfake / AI-generated / Manipulated)
- Class 1: Real (Authentic)

📈 Training Details

Dataset

Datasets: Utilizes over 50 diverse datasets, including selections from prithivMLmods, Kaggle, and other sources.
Content: Includes a wide variety of real photos and synthetic images, covering outputs from models like DALL-E 3, Imagen, and others.
Data Augmentation: Horizontal flips, rotations, color jittering

Hyperparameters

Parameter	Value
Base Model	google/siglip-so400m-patch14-384 (~400M params)
Fine-Tuned Model Size	~840 million parameters
Optimizer	AdamW
Learning Rate	5e-5
Loss Function	Cross-Entropy Loss
Epochs	10
Hardware	2x NVIDIA T4 GPUs
Training Time	~36 hours

✅ Evaluation Metrics

Metric	Score
Accuracy	96%
Precision	98%
Recall	98%
F1-Score	97%

⚠️ Note: These metrics were computed on a dedicated validation set. Performance may vary depending on real-world input distribution and image quality.

📌 Intended Use

The model is intended for detecting whether an image is real (authentic) or fake (manipulated or AI-generated). It can be applied in various domains:

Social Media Moderation
Digital Forensics
Journalism & Fact-Checking
Authentication Systems
Research & Development

🛠️ How to Use

Requirements

pip install transformers torch pillow gradio

Code Example

from transformers import AutoImageProcessor, SiglipForImageClassification
from PIL import Image
import torch

model_name = "AirasInnovations/Satyadrishti-V1-LARGE"
model = SiglipForImageClassification.from_pretrained(model_name)
processor = AutoImageProcessor.from_pretrained(model_name)

def detect_authenticity(image):
    image = image.convert("RGB")
    inputs = processor(images=image, return_tensors="pt")
    with torch.no_grad():
        logits = model(**inputs).logits
        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
    labels = model.config.id2label
    return {labels[i]: round(probs[i], 3) for i in range(len(probs))}

You can also use it directly via Gradio interface as shown in the example script.

🆚 Comparison with Vatsav Deepfake Detection

Feature	Vatsav Deepfake Detection	Satyadrishti-V1 Large
Training Data Age	3 years old	Jan–April 2025
Evaluation Data	Kaggle dataset	Multiple datasets including prithivMLmods Deepfake-vs-Real-v2
Open Source?	❌ No	✅ Yes
Accuracy	75%	96%
Precision	75%	98%
Recall	100%	98%
F1-Score	~86%	97%

📚 Citation & Acknowledgments

We thank Google Research for open-sourcing the SigLIP architecture and Prithiv MLMods for providing updated datasets for training.

If you use this model in your work, please cite:

@misc{airas-satyadrishti-large,
  author = {AIRAS INC},
  title = {Satyadrishti-V1 Large: An Open-Source Image Classification Model for Deepfake Detection},
  year = {2025},
  publisher = {Hugging Face},
  journal = {Model Card},
  howpublished = {\url{https://huggingface.co/AirasInnovations/Satyadrishti-V1-LARGE}}
}

📬 Contact

For support, feedback, or collaboration opportunities, please visit AIRAS INC Website or reach out to us at [email protected].

AirasInnovations
/

Satyadrishti-V1-LARGE