Vector Embeddings Tools — Pinecone & Qdrant Integration
What are the Vector Embeddings Tools?
To support advanced Retrieval-Augmented Generation (RAG) and semantic databases, Langoedge provides native embedding tools for Qdrant and Pinecone.
These tools convert unstructured text strings (like chat logs, user queries, or document paragraphs) into high-dimensional vector representations using your graph's configured embedding models (e.g. OpenAI's text-embedding-3-small).
[!IMPORTANT]
Prerequisite: Active Database Connection Required
Before your AI agents can read or write vector embeddings, you must connect your Pinecone or Qdrant keys on the Connect Page.
Without active credential connections established, vector embedding operations will fail to compile or execute at runtime.
Available Embedding Tools
1. Qdrant Embeddings (create_qdrant_embedding)
Generates text vector representations and inserts them into your Qdrant collections.
| Parameter | Type | Required | Description |
|---|---|---|---|
text |
string |
Yes | The source text string to embed. |
collection_name |
string |
Yes | Target collection name inside your Qdrant instance. |
2. Pinecone Embeddings (create_pinecone_embedding)
Generates text vector representations and inserts them into your Pinecone indices.
| Parameter | Type | Required | Description |
|---|---|---|---|
text |
string |
Yes | The source text string to embed. |
index_name |
string |
Yes | Target index name inside your Pinecone instance. |