Hacker News new | past | comments | ask | show | jobs | submit login

Wonder what the compute cost for this ends up being. My guess is they will have to somehow rate limit it - I’m sure the rollout speed will also depend how much people end up using it. Maybe they will figure out some ad revenue - but I don’t think they can charge users for this.

Would love to know how much compute $ they’re willing to throw at this to attempt to topple Google. Probably whatever it takes (within reason), MS missed out on mobile and search - can’t afford that to happen again.

Over time the inference for these queries will get cheaper (FP8 and later 4bit precision, sparse/pruned weights etc)




Maybe searches just arent that unique and they will cache all but the long tail of answers



This will go up because they're telling everyone to search like this:

I need to throw a dinner party for 6 people who are vegetarian. Can you suggest a 3-course menu with a chocolate dessert?

As compared to the autocomplete feature which guided people into cached search terms


What's the decay?

How many searches are only searched 2 and 3 times?

I'd question if it's worth it to store the results an AI output for 3 years so that you don't need to regenerate it when someone searches it in the future.

For 1 - you're probably going to have a meaningfully better model every year...


Yes, there’s a graph of cache duration to hit ratio.

With no data or particular domain expertise, I’d bet a 24 hour cache would hit at least 30% of the time, and a 7 day cache 60% of the time.

The problem is that a lot of the misses will be for recent events. How do you know whether “how many 737 maxes have crashed” should be served with a cache that is 3 months old?


Not really possible to cache. The model takes as input the entire chat (or some version of personalized history i guess).

But it s more likely that bing will discriminate between simple "keyword queries" and elaborate quesitons




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: