Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

Cloudflare Vectorize

developers.cloudflare.com

  • Free plan
  • Web
  • API

What Cloudflare Vectorize is

Cloudflare Vectorize is a globally distributed vector database for building AI-powered applications with Cloudflare Workers. It supports semantic search, recommendations, classification, anomaly detection, and supplying context to large language models. Vectors can come from Workers AI models or be provided by platforms such as OpenAI. Workers connect to indexes through a Vectorize binding configured in Wrangler or the Cloudflare dashboard; a REST API can insert and query vectors. Metadata filters support indexed string, number, and boolean properties, with up to 10 metadata indexes per Vectorize index. An index can contain up to 20 million vectors, each with a maximum of 1,536 dimensions at 32-bit precision. Queries can return up to 50 results with values or metadata, or up to 100 without them. The free plan includes 30 million queried vector dimensions and 5 million stored vector dimensions per month. Beyond paid-plan included amounts, usage is charged at $0.01 per million queried dimensions and $0.05 per 100 million stored dimensions. Cloudflare directs users to its #vectorize Discord channel for questions and feedback.

Who it suits

It suits developers building AI applications with Cloudflare Workers who need vector search or retrieval. The REST API offers another way to operate on indexes.

Good to know, and what to watch

  • Supports several AI application use cases.
  • Accepts vectors from Workers AI or OpenAI.
  • Free plan includes monthly queried and stored dimensions.
  • No billing for data transfer or egress.
  • Hybrid search is not supported.
  • Sparse vectors are not supported.
  • Queries return at most 50 results with values or metadata.

Workers Paid

—

Workers Free

Free

Free plan · paid from $5/mo

The verdict

Vectorize offers a free allowance and integrates with Workers through bindings or a REST API. Check its query result limits and whether its supported vector and metadata options fit your application.

What the maker publishes

Access security
Cloudflare describes Workers bindings as a capability and API, with the underlying secret never exposed to Worker code. source
API access
Vectorize supports a REST API for operating on indexes, including inserting and querying vectors. source
Billing
Vectorize does not bill for CPU, memory, active index hours, or the number of indexes, and does not charge for data transfer or egress. source
Capacity
An index can contain up to 20 million vectors, with a maximum of 1,536 dimensions per vector at 32 bit precision. source
Cloudflare integrations
Vector search results can reference images in R2, documents in KV, and user profiles in D1. source
Embeddings
Vectors can come from Workers AI models or be supplied from platforms such as OpenAI. source
Founded
2009 source
Headquarters
San Francisco, California, United States source
Metadata filtering
Metadata filters support string, number, and boolean indexed properties, with up to 10 metadata indexes per Vectorize index. source
Product
Vectorize is a globally distributed vector database for building AI powered applications with Cloudflare Workers. source
Query limits
Queries can return up to 50 results with values or metadata, or up to 100 results without them. source
Support
Cloudflare directs Vectorize users to its #vectorize Discord channel for questions, discussion, bug reports, and feedback. source
Use cases
It supports semantic search, recommendations, classification, anomaly detection, and providing context to large language models. source
Worker access
Workers connect to an index through a Vectorize binding configured in Wrangler or the Cloudflare dashboard. source

Checked against the maker's pages on 3 Oct 2026.