News for hackers. Stuff that matters.

A comment on What if AI worked at 1.000.000 tokens per seconds?.

Re:What if AI worked at 1.000.000 tokens per seconds? by ed on Saturday October 03, 2026 @02:32AM (#49941846)
At 1m tps, LLM’s could generate UI’s in realtime (a 5k webpage would take 5ms). Applications will have every little in common with today’s stacks when that happens. (For starters, rendering, event handling and data storage could move to the context.)
[ Parent | on HN ]
Re:What if AI worked at 1.000.000 tokens per seconds? by echohive42 on Saturday October 03, 2026 @02:39AM (#49941879)
yeah very true. I asked Opus 5.5 max if even such a thing is possible and it thought so, albeit after a lot of hurdles, especially memory related ones being overcome.

it also thought it could happen by 2033

[ Parent | on HN ]
Re:What if AI worked at 1.000.000 tokens per seconds? by hsellentin on Saturday October 03, 2026 @02:42AM (#49941892)
well 60fps is 16ms frame time so maybe we could get that by 2029 :)
[ Parent | on HN ]
Re:What if AI worked at 1.000.000 tokens per seconds? by echohive42 on Saturday October 03, 2026 @02:45AM (#49941904)
I hope so :)
[ Parent | on HN ]
Re:What if AI worked at 1.000.000 tokens per seconds? by beej71 on Saturday October 03, 2026 @03:37PM (#49947092)
And yet somehow we all know that it's going to be laggier than ever. :)
[ Parent | on HN ]
Re:What if AI worked at 1.000.000 tokens per seconds? by echohive42 on Saturday October 03, 2026 @07:31PM (#49948835)
i hope not :)
[ Parent | on HN ]