3 ms·
Google has so many users that not even they have enough GPU's and TPU's to service them all. I personally think they should use their best models, and just mak
by londons_explore 4y ago
Google has so many users that not even they have enough GPU's and TPU's to service them all.
I personally think they should use their best models, and just make it trigger very rarely. For example, only ~once per week per user (ie. 0.3% of queries).
Use a tiny model over the input query to decide if LaMBDA will do a far better job than regular search results, and only trigger in those cases where it will most benefit the user to begin with.