DoctorGPT: Offline & Passes Medical Exams!
Siraj Raval Siraj Raval
767K subscribers
23,739 views
0

 Published On Aug 12, 2023

Introducing DoctorGPT! After applying fine-tuning, reinforcement learning, & compilation techniques to Meta's Llama2 model, I got some amazing results. DoctorGPT is a Large Language Model that can pass the US Medical Licensing Exam. This is an open-source project with a mission to provide everyone their own private doctor. DoctorGPT is a version of Meta's Llama2 7 billion parameter Large Language Model that was fine-tuned on a Medical Dialogue Dataset, then further improved using Reinforcement Learning & Constitutional AI. Since the model is only 3 Gigabytes in size, it fits on any local device, so there is no need to pay an API to use it. It's free, made for offline usage which preserves patient confidentiality, and it's available on iOS, Android, and Web. Pull requests for feature additions and improvements are encouraged. In this video, i'll go over every single detail of how i built it so that you can do the same yourself. Enjoy!

Code: (renamed to Dr Dignity)
https://github.com/llSourcell/Doctor-...

Join my AI community:
http://gptschool.app

Please Subscribe! That's what keeps me going.

Want more AI/ML education? Connect with me here:
X: https://x.com/sirajraval
Instagram:   / sirajraval  
Facebook:   / sirajology  
Linkedin:   / sirajraval  
Tiktok:   / sirajraval  

I Built a Trading bot with Bing and ChatGPT:
   • I Built a Trading Bot with Bing and C...  

I Built a Sports Betting Bot with ChatGPT:
   • I Built a Sports Betting Bot with Cha...  

I Built a Trading Bot with ChatGPT:
   • I Built a Trading Bot with ChatGPT  

Watch ChatGPT Build an AI Startup:
   • Watch ChatGPT Build an AI Startup  

Watch ChatGPT Build a Finance Startup:
   • Watch ChatGPT Build a Finance Startup  

Learn Machine Learning in 3 Months:
   • Learn Machine Learning in 3 Months (P...   Sign up for my AI Sports betting Bot, WagerGPT! (500 spots available):
https://www.wagergpt.co

show more

Share/Embed