Generative AI Benchmarks: Evaluating Large Language Models

There are many variables to consider when defining our Generative AI strategy. Having a clear understanding of the use case/business problem is crucial. However, a good understanding of benchmarks and metrics helps business leaders connect with this new world and its potential.

So whether you are intending to: 

• select a pretrained foundation LLM (like OpenAI's GPT-4) to connect via API to your project, 
• select a base open-source LLM (like Meta's Llama 2) to train and customize, 
• or looking to evaluate the performance of your LLM 

the available benchmarks are crucial and useful in this task. In this video we will explore a few examples.