Thread

1. jmiskovic · 1789557527 · focus · HN ↗

   Incredibly important research. We've reached the point where local LLMs are good enough! It takes less time for local model to take the first action on your task than it does for Claude to validate your login, put you into queue and start issuing the commands. Local models are persistent and 100% predictable unlike any cloud offering. It's better for the power system for the demand to be distributed. During the winter time the GPU also doubles as a 300W in-house heater. Not to mention avoiding personal data collection and re-selling.

   1. 18Deepnar · 1789635351 · focus · HN ↗

      yea i see the potential of local slms too, it wont replace the frontier llms but for particular aspects they are generally really good, and at a certain point models like qwen 3.8 if people are able to set up proper api providers for it it can be extremely cheap and helpful. I am also working on trying to fix the memory and low context issue of such slms so they can actually be used locally for real use cases and not just small work

Unofficial Hacker News client; not affiliated with Y Combinator.