DocsDatabases

Pinecone

A fully-managed, serverless vector database. Zero ops, pay per read/write, scales to billions of vectors.

At a glance

License

Proprietary / managed

Hosting

Cloud only

Default index

Serverless (HNSW-class)

Best for

Teams that want zero ops

Create an index

import { Pinecone } from "@pinecone-database/pinecone";

const pc = new Pinecone({ apiKey: process.env.PINECONE_API_KEY! });

await pc.createIndex({
  name: "docs",
  dimension: 1536,
  metric: "cosine",
  spec: { serverless: { cloud: "aws", region: "us-east-1" } },
});

Upsert and query

const index = pc.index("docs");

await index.upsert([
  { id: "doc-1", values: vector, metadata: { source: "intro.md" } },
]);

const { matches } = await index.query({
  vector: queryVector,
  topK: 5,
  includeMetadata: true,
});

When to pick Pinecone

Pick Pinecone when you want a credit card to replace a platform team. Serverless pricing is read/write-based, so spikey workloads stay cheap. Avoid it if you need on-prem or strict data residency outside AWS/GCP/Azure regions Pinecone supports.