Re:The threat is comfortable drift toward not understanding ...
by danielbln on Sunday April 05, 2026 @09:32AM
(#47649278)
Open weight models that run under your desk are not frontier model level, but they are getting closer. Improvements in agentic post training and things like TurboQuant mean that even if all frontier labs pull the plug tomorrow, we will still have agents to work with.
[ Parent
| on HN ]
Re:The threat is comfortable drift toward not understanding ...
by zozbot234 on Sunday April 05, 2026 @10:26AM
(#47649795)
TurboQuant is not a step change, it's more of a smaller incremental improvement to KV quantization, and possibly (unsure) to quantization more generally. I'm actually more positive about SSD weights offload, which opens up very large local models for slow inference (good enough for slow chat) to virtually any hardware or amount of RAM.