vision-language model

English

Noun

vision-language model (plural vision-language models, abbreviation VLM)

  1. (machine learning) A generative AI model capable of interpreting text, images and video.
    Coordinate terms: large language model, small language model
    • 2025 February 25, Rina Diane Caballar, Cole Stryker, “What are vision language models (VLMs)?”, in IBM Think‎[1]:
      Vision language models can recognize and classify objects within an image and provide contextual descriptions such as an object’s position relative to other visual elements.

Further reading