VLM (Vision Language Model)

Models that take images plus text and produce text — "look at this and tell me X".

Definition

VLMs are multimodal models specialized in vision + language. Used in OCR, chart reading, screenshot UIs. Worker tasks include image-caption rating, visual question answering.

Examples

  • GPT-4 Vision
  • Claude 4 image input
  • LLaVA
  • BLIP

Related terms

Looking for actual gigs that use VLM (Vision Language Model)? Browse all live listings or filter by $100+/hr.