Efficient multi-prompt evaluation of LLMs

Open in new window