Speed-ups for State-of-the-Art Generative AI Drug Discovery Applications

Explore cutting-edge techniques for accelerating generative AI applications in drug discovery and molecular design in this 57-minute video presentation. Dive into summaries of 2024 reviews on generative AI for drug discovery, and examine a recent paper featuring a transcriptional signatures generator variational autoencoder for pancreatic cancer research. Learn about rapid GenAI screening tools using Hugging Face pipelines, and discover methods to improve inference speeds with C++ techniques like Llama.cpp, TensorRT, and LibTorch. Compare the performance of four large language models using the Groq Cloud inference engine, with insights on generation speeds and output quality as judged by ChatGPT and Gemini. Access the accompanying notebook and results on GitHub for hands-on exploration of these state-of-the-art approaches in AI-driven drug discovery.

Speed-ups for State of the Art Generative AI Drug Discovery Applications