We work around the clock to provide timely support and good developer experience.

Support

TopK

TopK is a cloud-native database built for search. It comes with native support for vector search, keyword search (with BM25 scoring), flexible filtering, re-ranking, automatic embeddings, and more to help you deliver the best search experience.

Introduction

Install the TopK Python or JavaScript SDK

Installation

Regions

Get started with the TopK search database in under 5 minutes.

Quickstart (5m)

Semantic search

Vector search

Multi-vector search

Keyword search

True Hybrid search

Reranking

Consistency

Multi-tenancy

Upsert documents

Use the `query` function to search for documents.

Query documents

You can retrieve a set of specific documents from a collection by calling `client.collection(name).get([ids])`.

Get documents

Delete documents

Collections are the core data structure in TopK. They store documents and provide the interface for querying them efficiently.

Create a collection

You can list all collections in a project by calling `client.collections().list()`.

List collections

You can get a specific collection in a project by calling `client.collections().get(name)`.

Get collection

Delete a collection

TopK is a cloud-native database built for search. It comes with embeddings, keyword search, vector search, metadata filtering, and reranking built-in.

Get started with the `topk` search database in under 5 minutes.

We work hard to provide a great support and developer experience. With that it is our aim to build a community that is a friendly place where people can hang out, help, and be helped. We are all here trying to build great experiences.

Community

Semantic Search

Unified Retrieval

Use the `query` method to search for documents.

Use the `upsert` method to update or insert documents.

Collections are the core data structure in TopK. They are used to store and query documents.

Get Started

Guides

Document API

Collection API

Support