# Quantization

Concept · [Building with AI](https://ecosystem.samcarlton.com/ai)

Also called: quantized, quantize, quantised, 4-bit, 8-bit, GGUF

Shrinking a model by storing its [weights](https://ecosystem.samcarlton.com/weights) with fewer digits of precision. The model gets smaller and faster, usually with a small [loss](https://ecosystem.samcarlton.com/loss-function) in quality, which helps run it on laptops and phones.

**For example:** A model too large for a laptop runs comfortably after it is quantized to 4-bit [weights](https://ecosystem.samcarlton.com/weights).

## Connects to (5)

- [Weights](https://ecosystem.samcarlton.com/weights): The numbers inside a model that it learned during training.
- [Local model](https://ecosystem.samcarlton.com/local-model): A model that is downloaded and run on a person's own computer or a company's own servers, not on a provider's.
- [Apple Silicon](https://ecosystem.samcarlton.com/apple-silicon): Apple's own family of chips used in Mac computers and iPads.
- [Distillation](https://ecosystem.samcarlton.com/distillation): Training a smaller model to copy the outputs of a larger one, so it keeps much of the quality while being cheaper and faster to run.
- [LoRA](https://ecosystem.samcarlton.com/lora): Low-rank adaptation.

Map view: https://ecosystem.samcarlton.com/quantization