SvectorDB Serverless vector database built for AWS

Usage Based

Developer Tools

What is SvectorDB?

SvectorDB is a serverless vector database solution specifically engineered for AWS environments. The platform enables efficient vector search capabilities while maintaining a cost-effective approach through its pay-per-use pricing model, eliminating the need for upfront costs or minimum fees.

The service features built-in vectorizers for text and images, hybrid search capabilities using Lucene/ElasticSearch style queries, and instant updates for upserts and deletions. With native CloudFormation support and the ability to handle up to 1 million vectors per database, SvectorDB provides a robust infrastructure for vector search operations while maintaining consistent performance.

Features

Hybrid Search: Filter results using Lucene/ElasticSearch style queries
Instant Updates: Real-time reflection of upserts and deletions
Built-in Vectorizers: Support for text and image vectorization
CloudFormation Support: Integration with existing AWS templates
Pay-per-request Pricing: No minimum fees or upfront costs
Multi-language Support: JavaScript, Python, and OpenAPI compatibility

Use Cases

Recommendation Engines
Document Search
Image Search
Retrieval Augmented Generation

FAQs

What are the current limitations of SvectorDB?

SvectorDB has a default limit of 1 million records per database, does not offer user-created snapshots, and is run by a micro start-up team.
What is included in the free tier?

The free tier includes up to 10 free tier indexes of up to 5k records each, with no time limit.
What are the pricing components?

Pricing includes storage ($0.25/GB/month), queries ($5/million), and writes ($20/million).