AI Dictionary › AI Models
GPT, Generative Pre-trained Transformer, identifies OpenAI's family of language models and, by extension, the architecture that enabled the current AI generation. The three terms describe how it works: Generative (it generates text), Pre-trained (pre-trained on enormous corpora), Transformer (the underlying neural architecture).
The GPT series has marked recent AI history: GPT-2 (2019) showed its potential, GPT-3 (2020) made text generation credible, GPT-4 (2023) brought reasoning and multimodal capabilities, and later versions pushed further on speed, reasoning and native voice and vision integration.
"GPT" has also become a generic term: OpenAI's "GPTs" are custom assistants built on top of the models, and many people say "a GPT" to mean any language model, even though Claude, Gemini and Llama technically do not belong to the GPT family.
From our network
AGORÀ Intelligence: Enterprise AI Governance Platform
Govern AI at scale: policies, adoption and measurable results on your data. Built for boards and C-suite.
Visit agora-intelligence.com →From the Agora Intelligence blog
📱 Download the Android app (beta) iOS coming soon
Say what you mean. Get what you need.
Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.