We needed a better way to choose LLMs

Show HN: We needed a better way to choose LLMs

We needed a better way to choose LLMs operates in show hn launch.

Profile

One of our team members — a backend dev — was trying to automate internal tasks using LLMs. He spent 2+ weeks researching, testing, and comparing models. No ML background. Just pure beast mode brute forcing it.Issue came clear atp: “Which LLM is actually suitable for this task?"Benchmarks didn’t help. Most are academic, vague, or not intuitive with real tasks.What we really wanted to know was: “For my specific use case, which model performs best — not in theory, but in practice?” So we built MIOSN. It stands for: Model, Input, Output, Score, Nominate — basically the way we select LLMs. With MIOSN, you: Define your own task (real input/output examples)Set what matters most (accuracy? cost? JS

The company sits in the show hn launch segment, where it competes on its core product or service.

It is recorded at the Show HN launch stage.

Website: miosn.com