Tutorial on how to use ChainForge to compare between different prompts, using template chaining. No cuts, no script --everything you see is live. Check out the final flow that appears in the video here: https://shorturl.at/fpNY5
ChainForge (chainforge.ai) is an open-source visual programming environment for prompt engineering, LLM evaluation and experimentation. With ChainForge, you can compare across prompts and text generation models with little to no coding required.
Note that this is just one way to compare across prompts. You could also set up evaluations to score responses automatically, and then visualize the results. However, I find that manual inspection is the best approach for early-stage prompt engineering.
GitHub: https://github.com/ianarawjo/ChainForge
Docs: https://chainforge.ai/docs/
Keywords: prompt engineering, LLMOps, LLM engineering, prompt comparison, ChainForge, machine learning, AI, ChatGPT, GPT4