The Pinecone Vector Database

Storing and searching through AI representations of your data lets you build search apps that “get it” and chatbots that answer right. Without exposing your data or fine-tuning models.

What is a vector database

High Performance

Build fast, accurate applications powered by Pinecone's proprietary algorithms and indexes optimized for throughput.

Real-time queries

Support for ultra-low query latencies even across billions of vectors.

Fresh, relevant results

Get relevant results with live index updates, filters, and hybrid search support.

  • title

    Introducing support for sparse-dense embeddings for better search results

    Our index now supports sparse-dense embeddings regardless of the model used, data type they represent, or data distribution. More ways to build means engineers can focus more on building great search applications.

    Read the Blog Post
  • title

    Performance Tuning

    This section provides some tips for getting the best performance out of Pinecone, reducing latency and increasing throughput.

    Read Guide
  • title

    The Missing WHERE Clause in Vector Search

    Pinecone’s single-stage filtering produces the accurate results of pre-filtering at even faster speeds than post-filtering.

    Read the Article

Effortless Operations

Choose a cloud region and we’ll do the rest. Get enterprise-level support, reliability, and security. Scale with zero effort and zero downtime. 

Low operational cost

Start for free and seamlessly scale with transparent, resource-based pricing that puts you in control. Easily monitor usage and streamline procurement through cloud marketplaces.

Zero-downtime scaling

With vertical and horizontal scaling, we can support billions of vectors and up to thousands of queries per second without interrupting application operations.

Fully managed

Choose the cloud provider and region that works best, and we’ll handle the rest. We support AWS, GCP, and Azure.

Google Cloud PlatformAmazon Web ServicesMicrosoft Azure

Secure by design

GDPR-ready, SOC2 Type II certified, HIPAA compliant. Easily control and manage access within the console with organizations and SSO. Data is encrypted at rest and in transit.


Easy to Use

A few clicks and you're up and running. Extensive developer library to jumpstart building and integrations to connect your stack.

Start in Seconds

Create an account and your first index with a few clicks or API calls. Become an expert with extensive developer docs, example notebooks, and educational articles.

Choose your AI model

Compatible with embeddings from any AI model or LLM, including those from OpenAI, Anthropic, Cohere, Hugging Face, PaLM, etc.

Simple APIs and SDKs

Intuitive REST API and growing number of SDKs make working with Pinecone a breeze for developers.