Pinecone is a managed vector database for AI applications. Use Stepper to upsert, query, and manage vectors for semantic search, retrieval-augmented generation (RAG), and recommendation workflows.
Insert or overwrite one or more vectors in a Pinecone index. Accepts vectors with their pre-computed embeddings and optional metadata.
3 parameters
Index
Namespace
Vectors
Query Vectors
Search a Pinecone index for the most similar vectors to a query vector. Supports metadata filtering.
8 parameters
Index
Namespace
Query Vector
Query Vector ID
Top K
Metadata Filter
Include Metadata
Include Values
Fetch Vectors
Retrieve one or more vectors by ID from a Pinecone index.
3 parameters
Index
Namespace
Vector IDs
Update Vector
Update the values or metadata of a single vector by ID in a Pinecone index.
5 parameters
Index
Namespace
Vector ID
Values
Set Metadata
Delete Vectors
Delete vectors from a Pinecone index by ID, by metadata filter, or delete all vectors in a namespace.
5 parameters
Index
Namespace
Delete Mode
Vector IDs
Metadata Filter
Generate Embeddings
Generate vector embeddings for one or more text inputs using a Pinecone-hosted embedding model.
4 parameters
Model
Text Inputs
Input Type
Truncate
Upsert Text (Embed + Upsert)
Generate an embedding for a text input using Pinecone's inference API and upsert it into an index in a single action. Ideal for populating a Pinecone index from a workflow without running your own embedding service.
6 parameters
Index
Namespace
Embedding Model
Vector ID
Text
Metadata
Rerank Documents
Reorder a list of documents by relevance to a query using a reranking model.
4 parameters
Model
Query
Documents
Top N
Create Index
Create a new serverless Pinecone index.
6 parameters
Name
Dimension
Distance Metric
Cloud
Region
Deletion Protection
Make HTTP Request
Make an HTTP request to any URL with full control over method, headers, and body.