HackerTrans
TopNewTrendsCommentsPastAskShowJobs

Readerium

no profile record

comments

Readerium
·10 days ago·discuss
3.5 Pro not yet launched, you mean 3.1 pro?
Readerium
·last month·discuss
Loll
Readerium
·2 months ago·discuss
For MSFT: Just download DeepSeek locally and use it.

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

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