VRAM stands for video random-access memory. It is memory used by your graphics processor (GPU) to hold data needed for tasks such as rendering graphics and running AI models.
For local AI, VRAM capacity helps determine which models and workloads your GPU can handle. Model weights, working data, and intermediate results all need memory. Larger models, higher image resolutions, and longer context lengths can increase that demand.
When a workload exceeds available VRAM, some tools can move part of it into system RAM or use memory-saving settings. This can reduce speed. Other tools may stop with an out-of-memory error.
VRAM is one part of hardware planning. GPU performance, system RAM, software support, and the requirements of your chosen model also matter.
