Marvin is an AI toolkit for developers who want to use LLMs with traditional software. We still see significant challenges integrating LLMs because of how difficult it is to get them to reliably accept and return structured data. Marvin consists of independent, functional tools that address this problem in a variety of ways.
Marvin has always been focused on using LLMs to work with native Python datatypes and Pydantic models. In 2.0 we've expanded this significantly with dedicated APIs for the most common use cases we've seen over the last year: classification, entity extraction, transforming data to types, and generating synthetic data. Marvin 2.0 is also fully multi-modal and supports images as inputs for classification, extraction, and transformation tasks (as well as simple image and speech generation). We've also introduces a Pythonic interface to OpenAI's assistants API, which now powers all of Marvin's interactive components.
We've tried to make an LLM framework that "sparks joy" and captures that same feeling you had the first time you saw an LLM in action. Try it out and let us know what you think!
(Repo: https://github.com/PrefectHQ/marvin) (Previous Show HN: https://news.ycombinator.com/item?id=35366838)
