AI news
公司内部站点 · 页面资源保存在本地(2026-10-19 16:40 UTC)· 独立 X 采集尚未配置;互动计数仅限本站

Bailey Jennings compares Jev and Luna on ONET job classification

Bailey Jennings

Bailey Jennings ran a quick demo of TypeSafe AI's Jev against Luna on ONET job classification. Luna reached 84.9% exact, about 2 seconds per job, and $482 per 1M. Jev reached 80.5% exact, about 200 milliseconds per job, and $122 per 1M.

He summed up the difference as roughly 10 times faster and 4 times cheaper for about 4 points of accuracy.