Digist
Detail

Show HN: JevBench, a reproducible benchmark for typed decision models

Original · Hacker News: Front Page · 26d ago
Hi HN! I built JevBench because Jev kicks ass, and the world deserves to know how the serious open source and fake lookalike projects really perform in comparison.Jev-class models return bounded choices and probabilities instead of text, and are disruptively faster and cheaper than LLMs, while being similarly intelligent on the text input they operate on.JevBench allows looking at accuracy, latenc…

Please improve

Tap what is wrong. A note is optional.