SOTA for excel generation

We're rolling out a feature that lets you vibecode Excel models, the same category a well-funded Silicon Valley startup has been getting attention for. Building it out, we realised we don't just match what's currently considered best. We beat it.

Here's our output for the prompt "build a detailed LBO model for ExxonMobil": https://docs.google.com/spreadsheets/d/1gsgJsKH8Mn5aepHipX9nP4-MDJ7lFT1j/edit?usp=sharing

Run that same prompt on any other platform and the gap is hard to miss.
Here is what claude had to say about our results vs others:
https://claude.ai/public/artifacts/35c88dcf-ef81-4ae5-8c98-5cd58f5d903b

How we did it: we generate Excel server-side instead of client-side, using openpyxl as the engine. We built on top of it to support pivot tables and everything else a real model needs. The reasoning behind the bet: openpyxl is heavily represented in training data, so the model actually knows how to drive it. That's the difference between a spreadsheet that looks right and one that is right.

What this unlocks: we are now very close to a finished AI associate. Associum already searches like Glean, executes tasks like Manus, builds decks like Gamma straight from raw data, and produces publication-ready reports that hold up against anything out there. Excel generation was the last big piece.

Give it a try. We are not based in Silicon Valley, so attention does not find us by default. Judge us on the output.