ChainForge is an open-source, visual programming environment focused on prompt engineering. It allows you to assess the robustness of prompts and text-generating models, going beyond simple case studies. We believe that testing multiple large language models, comparing their responses, and testing hypotheses about them should be not only easy but also fun. ChainForge provides a suite of tools to evaluate and visualize the quality of prompts (and models) with minimal effort. In other words, it aims to make the evaluation of large language models simple. ChainForge supports out-of-the-box testing for the robustness of prompt injection attacks, testing the consistency of response formatting, sending a large number of parameterized prompts and exporting them to Excel files, validating the response quality of the same model with different settings, and measuring the impact of different system messages on ChatGPT outputs, and more.