MetaCLIP-2-Age-Range-Estimator is an image classification vision-language encoder model fine-tuned from facebook/metaclip-2-worldwide-s16 for a single-label classification task. It is designed to predict the age range of a person from an image using the MetaClip2ForImageClassification architecture.

[!note] MetaCLIP 2: A Worldwide Scaling Recipe : https://huggingface.co/papers/2507.22062

Classification Report:
                  precision    recall  f1-score   support

      Child 0-12     0.9763    0.9758    0.9761      2193
  Teenager 13-20     0.9158    0.8437    0.8783      1779
     Adult 21-44     0.9593    0.9779    0.9685      9999
Middle Age 45-64     0.9458    0.9450    0.9454      3785
        Aged 65+     0.9769    0.9381    0.9571      1260

        accuracy                         0.9559     19016
       macro avg     0.9548    0.9361    0.9451     19016
    weighted avg     0.9557    0.9559    0.9556     19016

The model categorizes images into five age ranges:

Class 0: "Child 0-12"
Class 1: "Teenager 13-20"
Class 2: "Adult 21-44"
Class 3: "Middle Age 45-64"
Class 4: "Aged 65+"

Run with Transformers

!pip install -q transformers torch pillow gradio

python

import gradio as gr
import torch
from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image

# Model name from Hugging Face Hub
model_name = "prithivMLmods/MetaCLIP-2-Age-Range-Estimator"

# Load processor and model
processor = AutoImageProcessor.from_pretrained(model_name)
model = AutoModelForImageClassification.from_pretrained(model_name)
model.eval()

# Define labels
LABELS = {
    0: "Child (0–12)",
    1: "Teenager (13–20)",
    2: "Adult (21–44)",
    3: "Middle Age (45–64)",
    4: "Aged (65+)"
}

def age_classification(image):
    """Predict the age group of a person from an image."""
    image = Image.fromarray(image).convert("RGB")
    inputs = processor(images=image, return_tensors="pt")

    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits
        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()

    predictions = {LABELS[i]: round(probs[i], 3) for i in range(len(probs))}
    return predictions

# Build Gradio interface
iface = gr.Interface(
    fn=age_classification,
    inputs=gr.Image(type="numpy", label="Upload Image"),
    outputs=gr.Label(label="Predicted Age Group Probabilities"),
    title="MetaCLIP-2 Age Range Estimator",
    description="Upload a face image to estimate the person's age group using MetaCLIP-2."
)

# Launch app
if __name__ == "__main__":
    iface.launch()

Sample Inference:

Screenshot 2025-11-13 at 01-14-28 MetaCLIP-2 Age Range Estimator

Screenshot 2025-11-13 at 01-15-41 MetaCLIP-2 Age Range Estimator

Screenshot 2025-11-13 at 01-17-31 MetaCLIP-2 Age Range Estimator

Screenshot 2025-11-13 at 01-18-15 MetaCLIP-2 Age Range Estimator

Screenshot 2025-11-13 at 01-18-52 MetaCLIP-2 Age Range Estimator

Intended Use:

The MetaCLIP-2-Age-Range-Estimator model is designed to classify images into five age categories. Potential use cases include:

Demographic Analysis: Supporting research and business insights into age distribution.
Health and Fitness Applications: Assisting in age-based health recommendations.
Security and Access Control: Enabling age verification systems.
Retail and Marketing: Enhancing personalization and customer profiling.
Forensics and Surveillance: Supporting age estimation in investigative and security contexts.