Vision Model
Domain: Configuration
Motivation
An auxiliary model used for image analysis. Configured in model.vision.
Concrete Example
model.vision: openrouter/openai/gpt-4o. You send a photo. The vision model analyzes it: This is a cat sitting on a desk. The main model uses this description.
Analogy
A translator for sign language - they watch the signs and describe them in words. The vision model watches images and describes them in words.