Whilst I do like reading these things for technical know how, I can sympathise with the creator of jev who now presumably has to apply an order of magnitude effort to explain why the 100 smaller things done better than this add up to a much better product.
Except in this case it's not technically correct. Jev's claim is that it's frontier intelligence and these guys are pretending that a 8-bit quantized 0.6B param Qwen model is that. There's no universe in which that claim is technically correct.
dhsysusbsjsi · · focus · HN ↗
ramon156 · · focus · HN ↗
While technically correct, it's not the same thing
baobabKoodaa · · focus · HN ↗
hartator · · focus · HN ↗
baobabKoodaa · · focus · HN ↗