This video presents a comparative analysis of several GPT-based AI models, evaluating their performance across three distinct problem domains: mathematics, C++ programming, and electrical network solutions. The AI models included in the study are ChatGPT, Gemini, GitHub Copilot, Perplexity, and Claude AI. Each model was scored based on its ability to generate accurate and efficient responses to the given tasks.
The programming task specifically involved CUDA programming, with the base code drawn from the following NVIDIA documentation: https://docs.nvidia.com/cuda/cudss/in….
The detailed scoring results can be accessed here: https://docs.google.com/spreadsheets/…
CHAPTERS
_______________
SPEAKERS
Sidath Gajanayaka
Attorney-at-Law, Software Engineer and Lecturer in Law
Co-Founder and Director – Digico Labs (Private) Limited
LL.M., LL.B., B.Sc.,
Ph.D. (reading) in Law
Devin Aluthge
Electrical Engineer, Associate Engineer – IESL, Former Lecturer.
M.Sc., B.Sc.,
Ph.D. (reading) in Electrical Engineering
