Repolode
Submit
pinecone-io

pinecone-io/pinecone-ts-client

Official TypeScript/Node.js SDK for Pinecone, a managed vector and full-text search database.

View on GitHubWebsite
Stars
279
Forks
50
Open issues
20
Last commit
2 days ago

What is it?

This is the official TypeScript and Node.js client for Pinecone. It lets server-side applications create and manage indexes, store documents and raw vectors, and run full-text, vector, and metadata-filtered searches against a Pinecone index.

The SDK supports three workflows: document storage with full-text search, bring-your-own-vector upserts and queries, and integrated embedding where Pinecone generates embeddings from document text using hosted models. It also exposes inference endpoints for embedding and reranking models, and clients for the Pinecone Assistant API, which provides chat over uploaded files.

The package requires Node.js 22 or later and TypeScript 5.2 or later. Because it uses the API key directly, it is intended for server-side use only and should not be bundled into browser code.

Key features

  • 01Store documents and run full-text search across multiple fields with filtering
  • 02Upsert, query, fetch, update, and delete vectors with metadata filtering
  • 03Create serverless indexes and manage existing pod-based indexes
  • 04Generate embeddings server-side with Pinecone's hosted models via integrated embedding
  • 05Standalone inference API for embedding and reranking models
  • 06Generic TypeScript types for typed metadata

Use cases

  • Adding semantic search to a Node.js backend over your own documents
  • Building retrieval-augmented generation (RAG) pipelines that store and query embeddings
  • Combining keyword search on text fields with vector similarity in one query
  • Building a recommendation system based on vector similarity

Quick start

  1. 1

    Install the SDK from npm

    npm install @pinecone-database/pinecone
  2. 2

    For Node projects using TypeScript, install Node type declarations

    npm install --save-dev @types/node
  3. 3

    Set your PINECONE_API_KEY and run the full-text search quickstart code from the README

Things to know

  • Intended for server-side use only; using it in a browser can expose your API key.
  • Requires a Pinecone account and API key, and Node.js 22 or newer.

Topics

#vector-database#semantic-search#full-text-search#typescript#nodejs#rag#embeddings#llm#pinecone#similarity-search

FAQ

What is pinecone-ts-client?
Official TypeScript/Node.js SDK for Pinecone, a managed vector and full-text search database. This is the official TypeScript and Node.js client for Pinecone. It lets server-side applications create and manage indexes, store documents and raw vectors, and run full-text, vector, and metadata-filtered searches against a Pinecone index.
Is pinecone-ts-client free to use? What license does it have?
pinecone-ts-client is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is pinecone-ts-client still maintained?
Actively maintained. The last commit to pinecone-ts-client was on Nov 5, 2026.
What can I use pinecone-ts-client for?
Adding semantic search to a Node.js backend over your own documents Building retrieval-augmented generation (RAG) pipelines that store and query embeddings Combining keyword search on text fields with vector similarity in one query Building a recommendation system based on vector similarity
How do I install pinecone-ts-client?
Install the SDK from npm: npm install @pinecone-database/pinecone For Node projects using TypeScript, install Node type declarations: npm install --save-dev @types/node Set your PINECONE_API_KEY and run the full-text search quickstart code from the README
Where can I get pinecone-ts-client?
The source code and installation instructions are on GitHub: https://github.com/pinecone-io/pinecone-ts-client

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects