fructose

Hey HN! With the release of OpenAI [Structured Outputs](https://openai.com/index/introducing-structured-outputs-in-t...) it gave me an idea to build a library to make it easier to build Ai powered with functions. It's somewhat inspired by the many cool libraries in Python. [Fructose](https://github.com/bananaml/fructose), [Ask Marvin](https://www.askmarvin.ai) and others but has a few main differences * For Typescript! * Chain AI functions easily and get type safety throughout the entire chain * Call typescript functions with natural language (or structured inputs). If you define an input schema for your function, it will map the natural language input to that input schema * Specify output schema for a function. When you call a typescript function you can specify the output schema. (I guess sort of like graphql but for functions?) Not sur

This approach may be too high-level "magic" to the point of being difficult to work with and iterate upon. Looking at the prompt templates (https://github.com/bananaml/fructose/tree/main/src/fructose/... ), they use LangChain-esque "just try to make the output to be valid JSON" when APIs such as GPT-4 Turbo which this model uses by default now support function calling/structured data natively and do a very good job of it (https://news.ycombinator.com/item?id=38782678), and libraries such as outlines (https://github.com/outlines-dev/outlines) which is more complex but can better ensure a dictionary output for local LLMs.

Hi HN! Erik here from Banana (formerly the serverless GPU platform), excited to show you what we’ve been working on next: Fructose Fructose is a python package to call LLMs as strongly typed functions. It uses function type signatures to guide the generation and guarantee a correctly typed output, in whatever basic/complex python datatype requested. By guaranteeing output structure, we believe this will enable more complex applications to be built, interweaving code with LLMs with code. For now, we’ve shipped Fructose as a client-only library simply calling gpt-4 (by default) with json mode, pretty simple and not unlike other packages such as marvin and instructor, but we’re also working on our own lightweight formatting model that we’ll host and/or distribute to the client, to help reduce token burn and increase accuracy. We figure, no t