3 ms·
If you have somewhere to put them, you can get 2 440W panels for ~$350 if you're okay with intermittent or mostly daytime use. Or for ~$130/kWh you can extend t
by ndriscoll 2mo ago
If you have somewhere to put them, you can get 2 440W panels for ~$350 if you're okay with intermittent or mostly daytime use. Or for ~$130/kWh you can extend that with batteries. Then you need an inverter or some kind of regulator, but all in your fully capitalized power is still less than an AMD or Intel GPU and a lot less than an nVidia GPU for home use (the context of this thread is how intelligence is not limited to datacenter deployments and can be done at home). Most home users probably aren't going to leave it running overnight all the time, so you really only need storage for morning/evening, maybe. If you had a 27B model on a cerebras-like chip, it'd probably be much faster than any human could interact with it, so it'd race to idle just like CPUs.
- eru 2mo ago> If you had a 27B model on a cerebras-like chip, it'd probably be much faster than any human could interact with it, so it'd race to idle just like CPUs. I don't think the main use case is for a human to directly interact with the raw token stream. You probably want reasoning and you want the thing to be able to program on its own. That uses way more tokens than you can read.