Foundation model
A large model trained on broad data that can be adapted to many tasks; the most capable are called frontier models.
Also: frontier model, large language model
A foundation model is trained once on very large, general data and then used for many purposes, such as chat, coding or image generation. Large language models such as GPT, Claude, Gemini, Llama and DeepSeek are the best known. "Frontier model" refers to the most capable models at a given time, which require the largest training clusters. Only a few organisations can afford to train them, so their compute plans drive much of the demand for neoclouds and hyperscale AI data centres.