News for hackers. Stuff that matters.

A comment on MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second. Parent comment

Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by mediaman on Monday June 08, 2026 @08:49PM (#48454663)
The funny thing about this comment is that neural networks are universal function approximators.

The most fundamental essence of what they do is exactly what you say they don't: estimate.

[ Parent | on HN ]
Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by airstrike on Monday June 08, 2026 @09:38PM (#48455073)
Funny and ironic in a way, but the point still stands that they do not actually estimate the time it will take.
[ Parent | on HN ]
Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by greenavocado on Monday June 08, 2026 @11:28PM (#48455985)
> they do not actually estimate the time it will take

You can't prove that )))

[ Parent | on HN ]
Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by airstrike on Tuesday June 09, 2026 @12:31AM (#48456427)
Right, but extraordinary claims require...
[ Parent | on HN ]
Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by greenavocado on Tuesday June 09, 2026 @10:55AM (#48461993)
Instructions unclear, hard drive reformat completed.
[ Parent | on HN ]