# Weights

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

Also called: model weights, weight, the weights, weights file, released weights

The numbers inside a model that it learned during training. They decide what the model outputs for a given input. When a model's weights are released, anyone can download the file of those numbers and run the model.

**For example:** A lab publishes the weights of a model. A [developer](https://ecosystem.samcarlton.com/software-engineer) downloads the file and runs the model on a company [server](https://ecosystem.samcarlton.com/server).

## Connects to (12)

- [Parameters](https://ecosystem.samcarlton.com/parameters): The values a model learns during training.
- [Model](https://ecosystem.samcarlton.com/model): The trained system at the center of an AI product.
- [Open source vs open weights](https://ecosystem.samcarlton.com/open-source-vs-open-weights): Two labels for AI models that people often mix up.
- [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.
- [Quantization](https://ecosystem.samcarlton.com/quantization): Shrinking a model by storing its weights with fewer digits of precision.
- [LoRA](https://ecosystem.samcarlton.com/lora): Low-rank adaptation.
- [Neural network](https://ecosystem.samcarlton.com/neural-network): A kind of software model made of many layers of simple math units connected by weights, loosely inspired by the brain.
- [Hugging Face](https://ecosystem.samcarlton.com/hugging-face): A platform where the machine learning community shares AI models, datasets and demo apps.
- [Parametric knowledge](https://ecosystem.samcarlton.com/parametric-knowledge): What a model knows from its training, stored in its parameters.
- [Gradient descent](https://ecosystem.samcarlton.com/gradient-descent): The method used to train neural networks: measure how wrong the model is, work out which direction to nudge each weight to be less wrong…
- [Backpropagation](https://ecosystem.samcarlton.com/backpropagation): The calculation that works out how much each weight in a neural network contributed to an error, layer by layer from the output back, so…
- [Llama](https://ecosystem.samcarlton.com/llama): Meta's family of AI models whose weights are published for download.

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