Replying to Avatar hodlbod

I'm pretty sure the benchmarks it advertises are manufactured as well. DeepSeek seems substantially dumber than other LLMs, just from anecdotal evidence.
Avatar
fiatjaf 11mo ago

I had that impression too, but maybe it matches the intelligence of AI lovers more, that's why they think it's so good.

Reply to this note

Please Login to reply.

Discussion

No replies yet.