AI engineering glossary · Generative media multimodal
What is image tokenizer?
Image tokenizer: A model, often a VQ-VAE or similar autoencoder, that converts an image into a grid of discrete tokens and can decode them back. It lets language-model-style transformers generate or read images as if they were text.
Image tokenizer: an example
A tokenizer turns a 256 x 256 image into 256 tokens, and a transformer predicts them one by one to generate a picture.
Related terms
More Generative media multimodal terms
- Image-to-image translation
- Inception Score
- Inpainting
- Late fusion
- Latent diffusion model
- Latent space
- LLaVA
- LoRA fine-tuning for diffusion
Where you learn it
Image tokenizer is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.