● Beta — KubeEra is in active development. Profiles are live; AI-perception measurement is rolling out.
KubeEra
W

Weaviate

App Definition & Delivery
Open source · BSD-3-Clause

Weaviate is an open-source vector database written in Go. It stores objects and their vector embeddings together, supports hybrid search (vector + keyword/BM25), and exposes GraphQL and REST/gRPC APIs.

updated 2026-07-01

What it is

Weaviate is an open-source vector database written in Go. It stores objects and their vector embeddings together, supports hybrid search (vector + keyword/BM25), and exposes GraphQL and REST/gRPC APIs. It's built for semantic search, RAG pipelines, recommendation systems, and classification workloads where you need approximate nearest neighbor search at scale alongside structured filtering.

It runs standalone, in Docker/Kubernetes, or as a managed cloud service (Weaviate Cloud). Multi-tenancy, replication, and horizontal scaling are built into the core rather than bolted on.

Who builds it and why

Weaviate is developed by Weaviate B.V., a company that sells a managed cloud offering and enterprise support on top of the open-source core. Founding details aren't publicly available, but the commercial model is the standard open-core pattern: OSS drives adoption, cloud/enterprise features fund development. 173 contributors and steady commit activity (last commit 2026-07-01) indicate active maintenance, not an abandoned side project.

It's listed under CNCF's App Definition and Development landscape category, but CNCF maturity level is not publicly available — treat it as landscape-listed, not as a graduated/incubating CNCF project with the governance guarantees that implies.

Production readiness signal

16,474 GitHub stars and 173 contributors is a healthy, established open-source project — not a hobby repo. BSD-3-Clause is a permissive license with no copyleft concerns for commercial use. Latest release v1.38.2 suggests a mature, iterating versioning cycle (past the pre-1.0 instability phase).

What's missing from this data: CNCF maturity tier, founding date, and any independent benchmarks on production incident history. If you're doing vendor risk assessment, you'll need to pull SLA history and incident postmortems directly from Weaviate Cloud's status page — not available here.

Who should use this

  • Teams building RAG applications who need hybrid search (vector + keyword) out of the box, not as a separate bolt-on.
  • Engineering teams already comfortable operating stateful services on Kubernetes who want a self-hosted vector DB with multi-tenancy and replication built in.
  • Organizations that need GraphQL as a first-class query interface alongside REST/gRPC.
  • Teams that want to avoid vendor lock-in to a single cloud's proprietary vector service (e.g., avoiding being tied purely to a hyperscaler's managed offering).

Who should NOT use this

  • Small teams with <100K vectors and simple similarity search needs — Postgres with pgvector will be less operational overhead and one less system to run.
  • Teams already deep in the Elasticsearch/OpenSearch ecosystem — their vector search add-ons may be "good enough" and avoid introducing a new datastore.
  • Anyone wanting a zero-ops, fully-managed-only experience with no self-hosting option ever in scope — while Weaviate Cloud exists, the OSS-first architecture means self-hosting is a first-class path, which is overhead if you never intend to use it.
  • Teams needing strict ACID transactional guarantees across vector and relational data — Weaviate is not a general-purpose OLTP database.
  • Shops that need a CNCF-graduated project specifically for procurement/compliance reasons — maturity tier here is not publicly confirmed.

Alternatives

  • Qdrant — Rust-based vector DB with a similar hybrid search feature set and comparable open-source licensing; often cited as Weaviate's closest direct competitor.
  • Milvus — CNCF-adjacent (graduated under LF AI & Data), built for large-scale deployments with a more complex operational footprint but proven at billion-vector scale.
  • pgvector (Postgres extension) — the right call if your vector needs are modest and you don't want to run a separate database system at all.

Pricing

Weaviate core is fully open source under BSD-3-Clause — no license cost, self-host it yourself on your own infrastructure. Weaviate Cloud (managed hosting) is a paid commercial offering on top; specific pricing tiers are not included in the data provided here — check Weaviate's pricing page directly for current rates.

Frequently asked

What is Weaviate?+
Weaviate is an open-source vector database written in Go. It stores objects and their vector embeddings together, supports hybrid search (vector + keyword/BM25), and exposes GraphQL and REST/gRPC APIs.
Who builds Weaviate?+
Weaviate is developed by Weaviate B.V., a company that sells a managed cloud offering and enterprise support on top of the open-source core. Founding details aren't publicly available, but the commercial model is the standard open-core pattern: OSS drives adoption, cloud/enterprise features fund development.
Is Weaviate production ready?+
16,474 GitHub stars and 173 contributors is a healthy, established open-source project — not a hobby repo. BSD-3-Clause is a permissive license with no copyleft concerns for commercial use. Latest release v1.38.2 suggests a mature, iterating versioning cycle (past the pre-1.0 instability phase).
Who should use Weaviate?+
Teams building RAG applications who need hybrid search (vector + keyword) out of the box, not as a separate bolt-on. Engineering teams already comfortable operating stateful services on Kubernetes who want a self-hosted vector DB with multi-tenancy and replication built in.
Who should not use Weaviate?+
Small teams with <100K vectors and simple similarity search needs — Postgres with pgvector will be less operational overhead and one less system to run. Teams already deep in the Elasticsearch/OpenSearch ecosystem — their vector search add-ons may be "good enough" and avoid introducing a new datastore.
What are the alternatives to Weaviate?+
Qdrant — Rust-based vector DB with a similar hybrid search feature set and comparable open-source licensing; often cited as Weaviate's closest direct competitor.
How much does Weaviate cost?+
Weaviate core is fully open source under BSD-3-Clause — no license cost, self-host it yourself on your own infrastructure. Weaviate Cloud (managed hosting) is a paid commercial offering on top; specific pricing tiers are not included in the data provided here — check Weaviate's pricing page directly for current rates.