Article complete

Fractal's Vaidya.ai Outperforms GPT-4, Achieves 83% in PG-NEET

The model has been trained on a vast dataset of over 850,000 images and text entries, sourced from open-source data, data annotation, data collection, and synthetic datasets.

Modern office space featuring seating area and a large display screen showing various images.

Key takeaways

  • Fractal's Vaidya.ai scores 83% on the PG NEET, outperforming existing AI healthcare models.
  • The platform offers free, multilingual access to personalized healthcare information and report interpretation.
  • Vaidya.ai aims to combat misinformation and enhance healthcare accessibility for non-English speakers.
  • Fractal's initiative seeks to bridge gaps in healthcare delivery and improve patient engagement.
  • Users can interact with Vaidya.ai using both text and images without any costs.

Takeaways drafted with AI from the article text.

Touted as the "ChatGPT for healthcare," Vaidya.ai aims to bridge the gap between consumers and accurate, personalised health information, especially amid the challenges posed by widespread misinformation, delays in doctor appointments, and limited consultation time.

Vaidya.ai has been tested rigorously, scoring 83% on the PG NEET, a national entrance exam in India for postgraduate medical programs. This performance places it among the top-performing AI models in the healthcare domain, outperforming all other models benchmarked by Fractal.

"The Fractal AI research team is dedicated to unlocking solutions to complex challenges and creating a meaningful impact for the world. vaidya.ai represents a significant step forward in making healthcare more accessible to everyone. Our goal with vaidya.ai is to enhance people's access by providing timely, reliable healthcare information," said Srikanth Velamakanni, Co-founder, Group Chief Executive & Vice Chairman, Fractal.

Vaidya.ai provides users with free access to interact with the model using both text and images, without any fees or subscriptions. The platform can interpret health reports, answer queries in multiple languages, and deliver context-specific information. For example, users can upload health reports and receive summaries in English, Hindi, and other Indian languages, making healthcare information more accessible to non-English speaking populations.

"vaidya.ai is designed to augment healthcare information and doesn't intend to replace medical advice from medical professionals," said Suraj Amonkar, Chief AI Research and Platforms Officer at Fractal. "The development involved training models and frameworks using LLM and VLM architectures and validating the models on 20+ years of PG NEET/ PG Medical examination papers, among other metrics to measure how well the models performed. While it offers valuable insights and information, it always encourages users to consult healthcare professionals for medical advice."

The model has been trained on a vast dataset of over 850,000 images and text entries, sourced from open-source data, data annotation, data collection, and synthetic datasets. This extensive training involved multiple Vision-Language Models (VLM) and Large Language Models (LLM), including a model with over 30 billion parameters, ensuring the platform delivers reliable and relevant responses.

Vaidya.ai is the latest addition to Fractal's portfolio of AI-driven platforms, following previous launches like marshallgoldsmith.ai, a coaching assistant, and kalaido.ai, a text-to-image generator available in 17 Indian languages.

Published

Spotted an error? Tell our editors. Read our editorial standards and corrections.

More in this story

Fractal · 5 stories
  1. Fractal Launches Flyfish, the First Generative AI Sales Platform
  2. Fractal Introduces Kalaido, New Indic Text to Image Diffusion Model 
  3. AWS and GenAI Help Fractal Analytics Reduce Call Handling Time by up to 15%
  4. You are here: Fractal's Vaidya.ai Outperforms GPT-4, Achieves 83% in PG-NEET
  5. India’s First AI Unicorn, Fractal Eyes $500 Mn IPO
About the author

Tech Journalist

Siddharth is a media graduate who loves to explore tech through journalism and putting forward ideas worth pondering about in the era of artificial intelligence.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.