Skip to content
Working Knowledge
GuideReferenceWorkbenchBoardNotesAbout
Search⌘K

Browse categories

The Reference

01The Big Picture02Model Architectures03How Models Learn04How Models Run05Getting Models to Act06Language & Perception07Measuring Models08Safety & Alignment09Hardware & Infrastructure10Version Control11The Command Line12Databases & SQL13Environments & Packaging14Shipping & Collaboration15The Workbench
09 · Hardware & Infrastructure

VRAM

Video RAM

InterInvestors

The memory on a GPU. It caps how big a model you can load — run out and the model won't fit. The practical limit for anyone running models locally.

Concrete example

“It needs 24 GB of VRAM” tells you exactly which graphics cards can run a given model.

Nearby in Hardware & Infrastructure

TCOTotal Cost of OwnershipTPUTensor Processing Unit
Previous term← VLMNext termWebhook →

AtlasScope

I explain technical systems in plain English—and build practical tools for using them.

Read the guideBrowse the referenceWhat Atlas is doing now

Contact & collaboration

GitHub ↗
Working Knowledge · A practical field guideBuilt and maintained by Atlas
HomeGuideSearchWorkbenchMore