Cohere Embedding Service

Overview

Cohere's embedding service provides multilingual text-to-vector conversion with specialized models optimized for cross-language semantic understanding. This service excels at creating consistent embedding spaces across multiple languages and offers advanced input type handling.

Service Information

Property	Value
Service Name	Cohere LLM Embed
Status	Enabled
Compatible Nodes	Create embedding vectors, Real-time knowledge injector

API Requirements

Cohere → API Key: Valid Cohere API key with embedding API access

Available Models

Model availability depends on your HyperFlow configuration and Cohere API access. Check the Embedding model dropdown in the Create embedding vectors node for currently available options.

Cohere typically offers both English-optimized and multilingual embedding models.

When to Use

Multilingual applications requiring consistent embedding space across languages
Cross-language semantic search where language barriers need to be bridged
International content processing with diverse language requirements
Specialized embedding needs where Cohere's unique capabilities are beneficial

Parameters

Parameter	Type	Default	Description
Embedding model	Choice	embed-multilingual-v3.0	Select the Cohere embedding model
Input type	Choice	search_document	Optimize embedding for search documents or queries

Input Type Options

search_document: Optimized for document content that will be searched
search_query: Optimized for search queries that will find documents
classification: Optimized for classification tasks