Optimizing Large Language Model Selection for Efficient GenAI Development

Learn how to effectively select and optimize Large Language Models (LLMs) for Generative AI applications in this comprehensive conference talk from the Toronto Machine Learning Series. Explore practical strategies for model selection as three Georgian engineers - Royal Sequeira, Aslesha Pokhrel, and Christopher Tee - demonstrate tools like DSPy and frugalGPT that streamline the decision-making process. Master techniques for evaluating different foundational and open-source models while considering crucial factors such as fine-tuning requirements, prompt engineering, cost implications, and performance metrics. Gain hands-on experience in implementing modular design approaches that simplify the complex process of choosing the most suitable LLM for specific use cases. Discover how to navigate the growing ecosystem of language models and make informed decisions that optimize both functionality and efficiency in GenAI development.