| Hi HN! Over the last several weekends, I've been building LLMFlows as an alternative to langchain. There's been a lot of discussion on the shortcomings of langchain in the past few weeks, but when I first tried it in March, I thought there are 3 main problems:
1. Too many abstractions
2. Hidden prompts and opinionated logic in chains which makes it hard to customize
3. Hard to debug This inspired me to try and build a framework that solves these 3 issues, and therefore I started building LLFlows with the "philosophy" of being "simple, explicit, and transparent." A few weekends later, I think I finally managed to reach a state where I feel it's ready to be shared. I would love to hear your feedback! Thank you! |
I frequently am putting together complex chains of prompts and found microsoft guidance to be quite nice. Have you looked at it?
It feels like the abstraction I'd like the most is some way to combine few shot prompting, voting and chain-of-thought.
I have to create a prompt chain like this fairly often: write a description for this thing, given these previous input and outputs as examples (few shot). Do it N times with high randomness, maybe with expert personas, then review the solutions one by one with pros and cons (chain of thought) and then use all that to create a better description (final answer).
Right now, I always have to write out and connect all the steps, but its fairly rote and I think other prompting chains have a similar repetitiveness to them.
https://github.com/microsoft/guidance