Weaviate
Open-source vector database with hybrid search, RAG and multi-tenancy built in.
Quick answer
Weaviate is an open-source vector database with hybrid search, RAG and multi-tenancy built in, listed under Data labeling & vector databases. A typical small-business use: vector store and hybrid search behind a company RAG or knowledge-base app. The project publishes no official RAM figure, so we do not estimate a monthly server cost. License: BSD-3-Clause (Community Edition) + Weaviate Enterprise License (wl/), which allows internal business use.
Last updated · facts verified against official docs on · sources
Activity data comes from GitHub and was last refreshed Oct 25, 2026.
What is Weaviate?
Weaviate stores objects and vectors together and serves semantic, keyword (BM25) and hybrid search from one API. It can vectorize at import via integrated model providers or accept pre-computed embeddings, and it includes generative search (RAG), multi-tenancy, replication and RBAC for production deployments.
Where it fits in a 10–200 person business
- Vector store and hybrid search behind a company RAG or knowledge-base app
- Semantic product, ticket or document similarity search on your own infrastructure
- Multi-tenant embedding store for SaaS features you host yourself
These use cases are our editorial suggestions, not claims made by the project.
Can a business use Weaviate? (license)
| License | BSD-3-Clause (Community Edition) + Weaviate Enterprise License (wl/) |
|---|---|
| Type | Open core: Core is MIT/Apache; enterprise folders need a paid license. |
| Internal business use | Allowed under the license text |
| What to watch | Source outside the wl/ directory is BSD-3-Clause and free for internal and commercial use. Code under wl/ needs a separate enterprise license key; the official Docker image can enable those features only with a key. |
This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.
Data & compliance: does Weaviate phone home?
| Telemetry (usage data sent to the vendor) | On by default By default Weaviate sends a daily payload with machine id, server version, host OS, modules used, object/collection counts, and (since v1.33, when available) cloud metadata such as AWS account ID, GCP project ID or Azure subscription ID. It does not send data objects or collection names. How to turn it off: Set DISABLE_TELEMETRY=true. |
|---|---|
| Other calls home by default (updates, marketplaces, default cloud services) | Yes, by default Telemetry posts outbound; cloud metadata is read from provider metadata endpoints when present. Model-provider modules you enable make their own API calls. How to turn it off: Set DISABLE_TELEMETRY=true; do not enable API-based modules you do not need. |
| Single sign-on (OIDC / SAML / LDAP) | Free By protocol: OIDC free. OIDC authentication is documented for self-hosted Weaviate and can be combined with API keys and RBAC. Source: docs.weaviate.io/deploy/configuration/authentication |
| Audit log | Not documented |
| Role-based access control (RBAC) | Yes, free Role-based access control is documented as a Weaviate Database feature for assigning roles and permissions to authenticated users. |
| Data residency / GDPR notes from the project | Not documented |
Checked against the project's own docs and repository on . "Not documented" means we could not find the answer in official sources; it does not mean the feature is missing or that no data is sent. Third-party services you connect (model APIs, vector databases, search engines) have their own data policies. Compare all apps.
What hardware does Weaviate need?
The project does not publish a minimum CPU/RAM figure, so we leave the official value blank rather than guess.
GPU: Optional. Needed only if you run GPU-accelerated local modules such as text2vec-transformers with ENABLE_CUDA=1; API-based vectorizers need no GPU.
No standalone server minimum is published. Smaller collections (under about 1M objects) do not need special resource planning; memory is driven by the HNSW index. The local Docker quickstart that also runs Ollama recommends at least 8 GB RAM, preferably 16 GB.
How much does it cost to host Weaviate per month?
The project publishes no RAM figure we could verify, so we do not estimate a cost.
How to deploy Weaviate
Method: Docker
# pin the image tag to a release from https://github.com/weaviate/weaviate/releases
docker run -p 8080:8080 -p 50051:50051 cr.weaviate.io/semitechnologies/weaviate:1.40.1
# or generate a docker-compose.yml with the interactive Configurator / sample compose in the docsFrom the official instructions: docs.weaviate.io/deploy/installation-guides/docker-installation. Always check the source for the current version before you run it.
Known gotchas when self-hosting Weaviate
- The default docker run enables anonymous access (AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED=true). Turn on API-key or OIDC auth before exposing the instance. source
- Telemetry is on by default (version, OS, module list, object counts, and on some clouds account/project IDs). Set DISABLE_TELEMETRY=true to opt out. source
- Enterprise Edition features under wl/ need a Weaviate license key; the BSD license does not grant those features or allow circumventing the key. source
- Memory must hold the HNSW index; plan RAM from vector count and dimensions, or use quantization / disk-based indexes for larger sets. source
Alternatives in Data labeling & vector databases
- Milvus Scalable open-source vector database for semantic search and RAG at large data volumes.
- Qdrant Fast vector database for semantic search and RAG back-ends.
- Chroma Open-source embedding database for vector, full-text and metadata search in RAG apps.
- Label Studio Data labeling tool for text, images, audio and video, used to build and evaluate AI models.
- pgvector Open-source PostgreSQL extension for vector similarity search, next to your relational data.
Weaviate FAQ
How much RAM does Weaviate need?
Weaviate does not publish an official RAM figure, so we do not state one. No standalone server minimum is published. Smaller collections (under about 1M objects) do not need special resource planning; memory is driven by the HNSW index. The local Docker quickstart that also runs Ollama recommends at least 8 GB RAM, preferably 16 GB.
Does Weaviate need a GPU?
Optional. Needed only if you run GPU-accelerated local modules such as text2vec-transformers with ENABLE_CUDA=1; API-based vectorizers need no GPU.
How much does it cost to self-host Weaviate?
The software is free to download. The project publishes no official RAM figure, so we do not estimate a monthly server cost.
Is Weaviate free for commercial use?
Weaviate is licensed under BSD-3-Clause (Community Edition) + Weaviate Enterprise License (wl/) (Open core). Internal business use is allowed under the license text. Source outside the wl/ directory is BSD-3-Clause and free for internal and commercial use. Code under wl/ needs a separate enterprise license key; the official Docker image can enable those features only with a key. This is a summary, not legal advice.
What are common problems when self-hosting Weaviate?
The default docker run enables anonymous access (AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED=true). Turn on API-key or OIDC auth before exposing the instance. Telemetry is on by default (version, OS, module list, object counts, and on some clouds account/project IDs). Set DISABLE_TELEMETRY=true to opt out. Enterprise Edition features under wl/ need a Weaviate license key; the BSD license does not grant those features or allow circumventing the key. Memory must hold the HNSW index; plan RAM from vector count and dimensions, or use quantization / disk-based indexes for larger sets.
Is Weaviate still maintained?
Yes, it is actively developed. Last commit: Oct 16, 2026. Latest release: v1.40.2 (Oct 22, 2026). Checked Oct 25, 2026.
How do I install Weaviate?
The official method is Docker. The exact commands are in the "How to deploy" section above and in the official instructions: https://docs.weaviate.io/deploy/installation-guides/docker-installation
What are self-hosted alternatives to Weaviate?
Other apps in Data labeling & vector databases: Milvus, Qdrant, Chroma, Label Studio, pgvector.
Sources
- GitHub repository: github.com/weaviate/weaviate
- Official documentation, Deploy instructions, Gotcha 1: docs.weaviate.io/deploy/installation-guides/docker-installation
- License text, Gotcha 3: github.com/weaviate/weaviate/blob/main/LICENSE
- Hardware requirements, Gotcha 4: docs.weaviate.io/weaviate/concepts/resources
- Gotcha 2: docs.weaviate.io/deploy/configuration/telemetry
- Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
- GitHub stars, commits and releases: GitHub REST API, fetched
Facts on this page were checked against the project's own repository and docs on 2026-10-25. Spotted something out of date? Tell us.