ModelRefs / Model Size — AI Glossary
Model Size — AI Glossary
The storage footprint of a model's weights, typically expressed in gigabytes; depends on parameter count and numerical precision. 5 GB.
Overview
A 7B parameter model at float16 (2 bytes/param) occupies ~14 GB. Quantization to INT4 reduces this to ~3.5 GB. Model size determines minimum VRAM for inference. GGUF files embed quantization metadata and are the standard format for local inference with llama.cpp and LM Studio.
Reference details
| Topic | architecture |
|---|---|
| Also known as | model footprint, VRAM requirement |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Model Size — AI Glossary.
Frequently asked questions
What is Model Size?
The storage footprint of a model's weights, typically expressed in gigabytes; depends on parameter count and numerical precision.
Is Model Size the same as model footprint?
Yes — model footprint, VRAM requirement are common aliases for Model Size.
What concepts are related to Model Size?
Closely related concepts include parameter count, quantization, gguf.