Note: This guide is intended to be used by users with basic technical knowledge about LLMs. If you are unfamiliar with this topic, we suggest liaising with someone more experienced to help you with the selection process.
The tool aims to provide helpful information on relevant evaluation methods, but cannot cover all possible scenarios and requirements. We recommend considering the details of your use-case and potential domain-specific evaluation methods beyond the recommendations provided by this guide.
Additionally, note that this guide only focuses on the technical evaluation of LLM outputs. Other important aspects of using LLMs in real-world systems, such as information security of the used infrastructure or human-computer interaction factors are out of scope of this tool and should be considered separately.