Enum EmbedTaskType

Google, Cohere Only. Task type for embedding content. It's in Google format, but it's converted to Cohere format with JsonConverter when using with Cohere provider.

Namespace: Glitch9.AIDevKit

Assembly: Glitch9.AIDevKit.dll

Syntax

[JsonConverter(typeof(EmbedTaskTypeConverter))]
public enum EmbedTaskType

Fields

Name	Description
Classification	Optimized for classification tasks. Generates embeddings suitable for input to classification models.
Clustering	Optimized for clustering large volumes of text. Use cases: topic grouping, data analysis.
Document	Optimized for embedding documents to be searched. Use case: storing chat session messages, notes, or document bodies for later retrieval.
SearchQuery	Optimized for embedding search queries. Use case: user inputs a search query like "AI agent storage methods" and you want to match it against documents. Characteristics: trained to produce embeddings suitable for short queries.
SemanticSimilarity	Optimized for computing semantic similarity between two texts. Use cases: question-answer matching, paraphrase detection, duplicate detection.
Unspecified	Default value. The model will choose the best type for the input content.