Embedding model
Turns text into a list of numbers that captures its meaning, so similar ideas can be found by math.
AI Model Types
Every major kind of AI model in plain English: what it does, how it works, when to use it, and open-source repos to try.
24 model types · Page 2 of 2
Reranker
Rescores a shortlist of search results by reading the query and each result together, for better ordering.
Encoder model (BERT-style)
Reads a whole text at once to understand it, ideal for classification, tagging and search rather than writing.
Mixture-of-experts model
A huge model where only a few specialist sub-networks run for each token, so it is big but fast.
State space model
An alternative to transformers that reads long sequences in linear time by carrying a compact running memory.
Vision transformer and image classifier
Recognizes what is in an image by treating it as a grid of patches, the backbone of modern computer vision.
Object detection and segmentation model
Finds where things are in an image, drawing boxes or exact outlines around each object.
Robotics and world model
Learns how the physical world responds to actions, so robots can see, plan and move.
Time series foundation model
Forecasts future values such as sales, traffic or demand from past numbers, often with no training on your data.
Classifier and decision model
Picks one answer from a fixed set of options, with a confidence score, instead of writing free text.
Reward model
Scores how good a response is, used to train and steer other models toward preferred behaviour.
Safety classifier
Checks prompts and responses for harmful content, so an AI app can block or flag them.
Each page explains what the model type does, how it works, what it is good and bad at, and links to open-source GitHub repositories you can try. For single terms, see the AI Glossary.