Mixture of Experts (MoE)
A model architecture where a router network selects a small subset of "expert" sub-networks per token, allowing much larger total parameter counts than are activated per forward pass. Powers frontier models like Mixtral, DeepSeek-R1, and reportedly GPT-4/5. Trades RAM cost for compute cost.
Related terms
- Large Language Model (LLM)
A type of AI model trained on vast amounts of text data to understand and generate human-like text. Examples include GPT-4, Claude, and Llama. LLMs power many GenAI applications.
- Reasoning Model
A class of LLMs (OpenAI o-series, Claude with extended thinking, DeepSeek-R1) trained to spend significant inference-time compute on internal reasoning before producing the final answer. Substantially better on math, coding, and multi-step problems at the cost of higher latency and token consumption.
Related on this site
Where this fits
Mixture of Experts (MoE) is part of the AI Fundamentals vocabulary used in the Generative AI Maturity Framework. See the full glossary for the complete set of 149 defined terms, or take the free maturity assessment to see where your organisation stands.