Back to Gap Map Infrastructure / Storage & Retrieval

Weaviate

open core / Overall score: 3.4

Weaviate is an open-source vector database that stores objects and vectors together, combining vector search with structured filtering, keyword and hybrid search behind a GraphQL and REST API. It runs embedded for development, self-hosted as a Docker or Kubernetes cluster, or on the vendor's hosted service, with optional modules that call out to embedding and reranking models.

No PyPI artifact is declared on purpose. weaviate-client (80M downloads a month, published from weaviate/weaviate-python-client) is the Python client for a self-hosted or hosted Weaviate instance, not the server; installing it gives you a way to reach a Weaviate, not a Weaviate. Declaring it would make the client's count the product's adoption on the next warehouse refresh, so the package is recorded in the resolution ledger instead and adoption reads from stars.

Openness

4 high confidence

4.0

license: BSD-3-Clause(OSI)
enterprise-dir: Weaviate-License
source: public
core features withheld: yes — Shard Self-Recovery, Namespaces and Deduplicated backups ship in the wl directory under the separate Weaviate License and need a license key to run

Weaviate's LICENSE splits the repository in two: code outside the wl directory is BSD-3-Clause, while code inside it is an Enterprise Edition available only under a separate commercial license and switched on by a license key. That functionality sits inside the published tree but outside the open terms, the same shape as Meilisearch's Enterprise Edition, rather than a hosted service sold beside a complete core.

https://api.github.com/repos/weaviate/weaviate recorded 2026-09-02
Repo metadata - license spdx_id BSD-3-Clause, private false, archived false, default branch main - for weaviate/weaviate.
https://docs.weaviate.io/deploy/enterprise recorded 2026-10-07
Enterprise Edition page: code in the wl/ directory "is licensed under the Weaviate License"; the features table lists Shard Self-Recovery, Namespaces and Deduplicated backups, with a "Without a license key" column saying what each refuses.
https://github.com/weaviate/weaviate/blob/main/LICENSE recorded 2026-10-07
LICENSE: outside the "wl" directory source code "is licensed under the BSD-3-Clause License, as defined in LICENSE-BSD"; within it, code "is only available under a separate enterprise license" and "features are enabled via license key".
https://github.com/weaviate/weaviate/blob/main/README.md recorded 2026-10-07
README License section: "Files in our wl/ folder are only available under a commercial license with a license key issued by us" and "Enterprise features require a license key."

Adoption

3 low confidence

3.0

Weaviate declares no package artifact: weaviate-client is a Python client for a running instance, not the server, so it is left out rather than counted and ignored. Weaviate deploys as a Docker or Kubernetes cluster with no published deployment count either, leaving GitHub stars as the adoption signal.

https://api.github.com/repos/weaviate/weaviate recorded 2026-09-02
Repo metadata - stargazers_count = 16,776 - for weaviate/weaviate.

Capability

4 medium confidence

4.0

Weaviate is a complete, distributed vector engine, and its optional vectorizer and reranker modules call out to an external or containerized model rather than hosting one inside the core serving engine, which is what keeps it one step below Vespa, the only system that evaluates models inside its own serving path.

https://github.com/weaviate/weaviate/blob/main/README.md recorded 2026-10-07
README documents a cloud-native vector database with horizontal scaling, replication, multi-tenancy and RBAC, and integrated vectorizer, reranking and generative modules that call external or containerized models rather than hosting inference in the core query engine.

Verified 2026-09-27