Recent Posts
I am done. It's over.
Substack
4Chan bypass?
/emacs/ general
Forking jschan to submit a PR for captcha logic
This is Huge
PRICE OF INTERNET IN 1995
Motorola and Graphene linked up
HAPPENING
CPchads
JEEFICATION OF GSOC
Minecraft source code leaked
where do i get CRT monitor ??
Shifting to linux mint
Androidfags zara idhar aana
Sarvam is now proven to be a disappointment
XHDATA D-808 DX-ing setup, analogue modulation
i don't understand
RCE on Pocketbase possible?
AI Impact Summit 2026
Simple Linux General /slg/ - Useful Commands editi...
/wpg/ - Windows & Powershell General
Sarvam Apology Thread
Some cool tech in my college
the hmd touch 4g
Holy Shit
Saar american companies have best privacy saaar
/g/ related blogpost - backup thread
Android Hygiene
My favorite game rn
COOKED
Are we getting real 5g?
I want to create my own forum but I don't know how...
Is my psu not compatible with my mobo?
/i2pg/ - I2P general
24/l2w
No.2983
How much electricity do you consume in your daily local model runs?
YMx7B4
No.2984
>>2983(OP)
I am gareeb, I don't have gpu and the one that I have in my laptop is gtx 1650 (which is good for general purpose, but not for running llm more that 2b).
yvCyyh
No.2985
>>2983(OP)
i dont use local llms as they are useless on my limited hardware
tGbf4Z
No.2986
> No GPU
> No life
tGbf4Z
No.2989
>>2983(OP)
I ran local models on my 13th i5. 65 watts peak. Ok for LLM use. Not fast. Image generation was not good.
I want to try to those new AMD AI cpus with npu's built into motherboard.
tGbf4Z
No.2990
>>2983(OP)
And while id love a 12 gb graphics कार्ड, thr costs even for a 3050 are high





















































