GRANITE_3_2_VISION
Represents the Granite 3.2 Vision model.
This predefined instance of LLModel is provided by Granite and supports the following capabilities:
Temperature: Allows adjustment of the temperature setting for controlling the randomness in responses.Schema.JSON.Simple: Supports tasks requiring JSON schema validation and handling in a simplified manner.Tools: Enables interaction with external tools or functionalities within the model's ecosystem.Vision.Image: Enables processing and understanding of image inputs alongside text.
The model is identified by the unique ID "granite3.2-vision" and categorized under the Ollama provider. It is designed to handle both text and visual inputs for multimodal tasks.