Hacker News Reader

They don't lead on latency nor quality; but their execution was superb

Who they trailing on quality?

I'm not the person you're asking, but:

According to this benchmark, Jev is currently trailing Quyet-1.0-Large and a few other hastily put-together LLM-based decision API-like setups.

And the 'better' ones are slower and cost more for a tiny bit more accuracy. Its hard to sell that as being better when speed and price have been JEVs main selling points.

The top alternative right now lists speed as faster than Jev?

The space of problems that require super low latency AND are inputting so many tens of billions of tokens that the price matters, is a tiny one.

Here's another leaderboard: https://huggingface.co/spaces/multimodalart/jev-decision-ind...

What even is this? What does it do?

Reply on news.ycombinator.com