Definition
The learned numbers inside a neural network. Training produces them; their count (e.g. 7B = 7 billion) signals a model's size, capability, and cost.
Weights are the millions or billions of adjustable numbers a model learns during training; they encode everything it knows. Parameter count is shorthand for scale — a 7-billion-parameter model is smaller, cheaper, and faster than a 70-billion one, usually at some accuracy cost.
Also known as
parameters, weights
Specialist software house for video, real-time and AI products. Founded 2005. 50 in-house engineers.
