SurrealDB review

The free tier is listed as available, so a small project can start without a paid plan. What that tier does not add is help: support runs on community only, and the engines in document, graph and vector are the same set you would grow into later.

The engine list is document, graph and vector, so document, graph, vector, time-series and relational models all run in the same database. That matters for teams that would otherwise run a separate document store, a graph engine and a vector index for one project.

Backups are part of the service as daily automated backups, so copies are taken on a routine schedule without a separate backup add-on. Teams that need a specific retention window or a tested restore plan should check that cadence against their recovery targets.

Support runs on community, which means the community is the channel listed. No email, chat or phone channel appears for the brand, so a team with a production incident and a response-time requirement should look at a vendor that lists those channels.

Across the database providers listed, the median entry price is €4.38, and the far end of entry pricing reaches €236.78. SurrealDB's entry point is a free tier, so the paid plan decision comes later.

Aiven lists Kafka, PostgreSQL and OpenSearch and Heroku lists PostgreSQL, Redis and Kafka, both sets of separate managed services. SurrealDB covers document, graph and vector inside a single product, so the choice is many models in one place against several engines managed side by side.

Vector is one of document, graph and vector, so vector search sits alongside document, graph, time-series and relational data in the same database. A team that only needs embeddings and similarity search can start on the free tier and keep the same models as usage grows.