Architecture
Vision-language model
A model that understands images together with text, answering questions about photos, charts, screenshots or documents.
Most frontier chat models are now vision-language models. They can read a scanned invoice, describe a diagram or reason about a screenshot, though fine detail and exact numbers still need checking.
Frequently asked questions
What does Vision-language model mean?
A model that understands images together with text, answering questions about photos, charts, screenshots or documents. Most frontier chat models are now vision-language models. They can read a scanned invoice, describe a diagram or reason about a screenshot, though fine detail and exact numbers still need checking.