Managed Vector Similarity Search

Search through billions of vector embeddings for similar matches, in milliseconds.
It’s the next generation of search, an API call away.

Pinecone similarity search as a service

By submitting, you accept the End-User License Agreement.

Want a demo or have questions? Contact us.

Why Pinecone?


Launch a distributed similarity search service in a few lines of code.

  • Create a service and start making API calls — leave the infrastructure and ops to us.

  • Each service is persistent, consistent, sharded, and replicated across many nodes.

Sample code for deploying Pinecone.
Sample code for indexing vectors.


Dynamically load and index billions of vector embeddings.

  • Load vector embeddings in streams or batches from models, data lakes, or feature stores.

  • Update from anywhere. New and updated items become searchable in milliseconds.


Run similarity search in Python or Java applications, or notebooks.

  • Latency of <50ms, even with billions of items and thousands of queries per second.

  • Our fast search algorithms find more accurate results than open-source options.

Sample code for searching vectors.


Pinecone runs on hardened AWS infrastructure. Data is stored in isolated containers and encrypted in transit.



Data persistence, eventual consistency, automatic node recovery, replication, sharding, and more. All done for you.



Only pay for what you use, as you go. The efficiency of our vector index leads to lower operating costs.

Ask us about VPC deployments on AWS or GCP.

What will you build?

Get started

Or contact us for a customized demo.