Transformer
Architecture / Model
Definition
A neural network architecture based on self-attention mechanisms that processes input data in parallel. Introduced by Google in 2017 (‘Attention Is All You Need’), the Transformer enabled scaling of language models and the generative AI revolution.
In French
Transformer — Architecture de réseau de neurones révolutionnaire basée sur l’attention, à la base de tous les LLM modernes. Introduit par Google en 2017 («Attention Is All You Need»), le Transformer a permis le passage à l’échelle des modèles de langage et la révolution de l’IA générative.
Related terms
Autoencoder
Diffusion Model
Foundation Model
GAN (Generative Adversarial Network)
Large Language Model (LLM)
Neural Network
🛠️ Related tools
Explore the full glossary
Discover all artificial intelligence terms in our glossary.