Mastodawn

RAM overflow even after closing all tasks #ram #thinkpad #memoryusage #2510 #ubuntubudgie

RAM overflow even after closing all tasks

The issue happens after a while of running ubuntu (Budgie), maybe after some suspend - waking up. Today it happened after about 5 minutes of using the firefox after waking up from suspend. Using the

Ask Ubuntu

AskUbuntu Apr 28

How to significantly reduce memory usage of GTK4 applications in computers with low RAM running Ubuntu? #ram #memoryusage #gtk4

https://askubuntu.com/q/1566199/612

How to significantly reduce memory usage of GTK4 applications in computers with low RAM running Ubuntu?

I use Ubuntu on a Chromebook with only 4GB of memory, and GTK4 apps consume a lot of memory. Examples: Even a hello world window may consume more than 100 mb of memory One app was reported to hav...

Ask Ubuntu

AskUbuntu Apr 14

Why is my Ubuntu VM running out of memory and closing unrar? #memoryusage

https://askubuntu.com/q/1565692/612

Why is my Ubuntu VM running out of memory and closing unrar?

My Ubuntu VM, running on TrueNAS, has 32GB RAM assigned, but cannot even unrar a file with nothing else running. System monitor says 97% or RAM is in use before I even run the command. After a min...

Ask Ubuntu

ResearchBuzz: Firehose Mar 26

Ars Technica: Google’s TurboQuant AI-compression algorithm can reduce LLM memory usage by 6x. “Google Research recently revealed TurboQuant, a compression algorithm that reduces the memory footprint of large language models (LLMs) while also boosting speed and maintaining accuracy.”

https://rbfirehose.com/2026/03/26/ars-technica-googles-turboquant-ai-compression-algorithm-can-reduce-llm-memory-usage-by-6x/

Ars Technica: Google’s TurboQuant AI-compression algorithm can reduce LLM memory usage by 6x

Ars Technica: Google’s TurboQuant AI-compression algorithm can reduce LLM memory usage by 6x. “Google Research recently revealed TurboQuant, a compression algorithm that reduces the memory fo…

ResearchBuzz: Firehose

aicoder Oct 27, 2025

#AIEngineering #memoryusage #cheatsheet #bestpractice

https://machinelearningmastery.com/3-ways-to-speed-up-model-training-without-more-gpus/

3 Ways to Speed Up Model Training Without More GPUs - MachineLearningMastery.com

Most training jobs run slower than they need to, not because the GPU is weak, but because it isn’t being used efficiently. This article shows three proven ways to speed up model training without adding more GPUs.

MachineLearningMastery.com