HackerLangs
TopNewTrendsCommentsPastAskShowJobs

Readerium

no profile record

comments

Readerium
·hace 10 días·discuss
3.5 Pro not yet launched, you mean 3.1 pro?
Readerium
·el mes pasado·discuss
Loll
Readerium
·hace 2 meses·discuss
For MSFT: Just download DeepSeek locally and use it.

Or train your own power efficient stack.
Readerium
·hace 2 meses·discuss
AI workloads are all about memory size and bandwidth not compute
Readerium
·hace 2 meses·discuss
LLMs are memory bandwidth bound not compute bound.
Readerium
·hace 3 meses·discuss
Add an iPad mini esque screen on the trackpad of a MacBook.

Trackpad of a 16 inch MacBook Pro is humongous anyways.

Add a touchscreen display to the trackpad, and give it iPad OS
Readerium
·hace 3 meses·discuss
that is true. gguf does not support any Architecture.

for the most recent example, as of April 16, 2026 (today)

Turboquant isnt still added to GGUF
Readerium
·hace 3 meses·discuss
perhaps increasing repitition_penalty might be helpful
Readerium
·hace 3 meses·discuss
In coding they are worse.

Chinese models (GLM, MiniMax) are better.
Readerium
·hace 4 meses·discuss
Can someone explain if the 3D Vcache are stacked on top of each other or side by side.

If they are stacked then why not 9800X3D2?
Readerium
·hace 4 meses·discuss
Qwen 3.5 4B is the goat then
Readerium
·hace 4 meses·discuss
Yes he is room-mates with.
Readerium
·hace 9 meses·discuss
99.999 percent lol
Readerium
·hace 10 meses·discuss
So was intel in the following sentence.
Readerium
·hace 10 meses·discuss
[dead]