HackerTrans
TopNewTrendsCommentsPastAskShowJobs

Readerium

no profile record

comments

Readerium
·10 dni temu·discuss
3.5 Pro not yet launched, you mean 3.1 pro?
Readerium
·w zeszłym miesiącu·discuss
Loll
Readerium
·2 miesiące temu·discuss
For MSFT: Just download DeepSeek locally and use it.

Or train your own power efficient stack.
Readerium
·2 miesiące temu·discuss
AI workloads are all about memory size and bandwidth not compute
Readerium
·2 miesiące temu·discuss
LLMs are memory bandwidth bound not compute bound.
Readerium
·3 miesiące temu·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
·3 miesiące temu·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
·3 miesiące temu·discuss
perhaps increasing repitition_penalty might be helpful
Readerium
·3 miesiące temu·discuss
In coding they are worse.

Chinese models (GLM, MiniMax) are better.
Readerium
·4 miesiące temu·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
·4 miesiące temu·discuss
Qwen 3.5 4B is the goat then
Readerium
·4 miesiące temu·discuss
Yes he is room-mates with.
Readerium
·9 miesięcy temu·discuss
99.999 percent lol
Readerium
·10 miesięcy temu·discuss
So was intel in the following sentence.
Readerium
·10 miesięcy temu·discuss
[dead]