Type: nodetool.agents.ImageAgent

Namespace: nodetool.agents

Description

Analyze and transform images with a vision model from a prompt. agent, image, vision, transform, analysis

Properties

Property Type Description Default
model language_model Model used for image prompts. {"type":"language_model","provider":"empty","id...
image list[image] Optional image inputs for image reasoning tasks. Accepts a list, or a single Image (auto-wrapped). []
prompt str Prompt describing the image task. ``
timeout_seconds int Maximum runtime for agent execution. 90
max_output_chars int Maximum serialized output chars before truncation. 120000

Outputs

Output Type Description
image image  
text str  

Browse other nodes in the nodetool.agents namespace.