I really like Promptfoo for testing prompts, so I wrote an article on how to use it to test prompts with different models and various assert types. Let me know what you think!

In the article, I show how to create evals with Promptfoo to test prompts like code. You can compare different models (open-source and proprietary) and use various assert types (equals, contains, g-eval, semantic similarity, JavaScript, etc.) to validate the output of your prompts.

https://pvkl.nl/en/automating-prompt-tests-with-promptfoo/