Parameters
The internal numbers a model learns during training, often in the billions. More parameters can mean more capability, but not always.
Parameter count is a rough size measure people cite, though quality of training matters just as much.
Frequently asked questions
Does more parameters always mean a better AI model?
No. Parameter count is a rough size measure, but the quality of the training data and method matters just as much. A smaller, well-trained model can outperform a larger, sloppily trained one on real tasks.
What are parameters in plain terms?
They are the internal numbers, often in the billions, that a model learns during training and then uses to make predictions. Think of them as the settings the model dials in as it studies patterns, not facts stored in a list.
Why do people mention parameter counts like 7B or 70B?
Those numbers describe how many parameters a model has, so they give a quick sense of its scale and roughly how much hardware it needs to run. They are a starting signal, not a guarantee of how well it performs for you.
New to all this? Start with what an AI agent really is, browse the full glossary, or explore the learning hub.
← Back to the glossary