Tootfinder

Opt-in global Mastodon full text search. Join the index!

No exact results. Similar results found.
@teledyn@mstdn.ca
2025-12-08 19:22:54

RE: cosocial.ca/@hyphacoop/1156853
They likely won't like my answers. 😅
FWIW a 7B param LLM costs around 55MWh to train and test, and about 300Wh per query. Then factor that 52% of the answers are unusable, so like Lay's potato chips, NOBODY does only one prompt.