baml

Hey HN, We made an LLM Playground where you can use jinja to render prompts and extract structured data. For example, receipts, invoices, audio snippets, etc. It uses the BAML programming language under the hood. I was tired of not being able to test every model (local + hosted) in the same place. I wanted it work for audio and for images and for text. Plus, I HATE navigating UIs when I can just write code, so the existing non-code first offerings sucked for me. We were inspired a little bit by Markdown Preview and a little bit by Rust. For example, the VSCode version of this has a run-tests buttons right in your IDE cause that is the only reason i started writing tests in rust. What do you think? Code is here: https://github.com/BoundaryML/baml/tree/canary/typescript

One of the issues with LLMs is that each function needs to be tested and observed more thoroughly than regular functions, which leads to these UIs to figure out what the heck is actually going on. For our DSL (BAML https://github.com/BoundaryML/baml ) we found that adding a VSCode playground to visualize LLM function inputs and outputs was a massive win in terms of debuggability and testability, so I can see why langraph is going this way.

thanks for the shoutout, we benchmarked our approach against other function-calling techniques and we've been able to beat all other approaches every time (even by 8%!) just by getting better at parsing the data and representing schemas with less tokens using type definitions instead of json schema. You can take a look at our BFCL results on that site or the github: https://github.com/BoundaryML/baml We'll be publishing our comparison against OpenAI structured outputs in the next 2 days, and a deeper dive into our results, but we aim to include this kind of constrained generation as a capability in the BAML DSL anyway longterm!