Weaviate

Vector storesBSD-3-Clause for most of the repository; files in the wl/ folder are commercial-only

Weaviate is an open-source vector database written in Go. It stores objects with vectors and answers vector, keyword and hybrid queries. You can run it in Weaviate Cloud, with Docker, on Kubernetes or embedded from Python or JavaScript. Official clients cover Python, TypeScript and JavaScript, Go, Java and C#, and the server also exposes GraphQL, gRPC and REST. From version 1.38 an MCP server is built in, disabled by default on self-hosted instances. Its tools read collection configuration, list tenants, run hybrid queries and upsert objects.

Pinout

  • MCP server: supported, verified
  • HTTP API: supported, verified
  • command line: not checked yet
  • Python SDK: supported, verified
  • TypeScript SDK: supported, verified
  • OpenAPI description: not checked yet
Where each pin comes from
InterfaceStatusSource we readRead on
MCP serverSupported, verifiedMCP server page
HTTP APISupported, verifiedClient libraries page, GraphQL, gRPC and RESTful APIs
Command lineNot checked yet——
Python SDKSupported, verifiedClient libraries page
TypeScript SDKSupported, verifiedClient libraries page
OpenAPI descriptionNot checked yet——

Facts

Facts about Weaviate
LicenseBSD-3-Clause for most of the repository; files in the wl/ folder are only available under a commercial license. GitHub repository, About sidebar and README, read .
Latest GitHub releasev1.40.5. GitHub releases, latest, read .
PlatformsWeaviate Cloud, Docker, Kubernetes, embedded launch from Python or JS/TS. Weaviate documentation home, read .
Client librariesPython, TypeScript/JavaScript, Go, Java, C#; GraphQL, gRPC and REST also available. Client libraries page, read .
MCP serverBuilt in since v1.38, runs at /v1/mcp, enabled with MCP_SERVER_ENABLED=true; always on in Weaviate Cloud. MCP server page, read .
Pricing, free plan$0/month, no credit card: 100,000 objects, 1 GB memory, 10 GB disk, 1 cluster per user. Pricing page, Free, read .
Flex planFrom $45/month, pay-as-you-go, no commitment. Pricing page, Flex, read .
Premium planPrepaid contract, minimum from $400/month per the FAQ. Pricing page, Premium, read .
Last commit. GitHub API, latest commit on the default branch of weaviate/weaviate, read .

Call example

docker run -p 8080:8080 -p 50051:50051 cr.weaviate.io/semitechnologies/weaviate:1.40.4

Limits

  • The MCP write tool (weaviate-objects-upsert) is only available when write access is enabled by a separate setting.

Strengths

  • The MCP server exposes a hybrid vector and keyword query tool.
  • The pricing page lists a Free plan.

Pricing

Pricing, free plan$0/month, no credit card: 100,000 objects, 1 GB memory, 10 GB disk, 1 cluster per userPricing page, Free, read 16 Nov 2026
Flex planFrom $45/month, pay-as-you-go, no commitmentRead 16 Nov 2026
Premium planPrepaid contract, minimum from $400/month per the FAQRead 16 Nov 2026

As written on Weaviate's own page on the read date. Prices change: check there before you pay.

Alternatives to Weaviate

All vector stores tools for agents
  1. Rust vector database with REST, gRPC, SDKs and an MCP server.

    Vector stores
  2. Embedded vector database library with Python, TypeScript and Rust SDKs.

    Vector stores
  3. Open-source vector database with Lite, Standalone and Kubernetes deployments.

    Vector stores
  4. Postgres extension that stores vectors and runs nearest-neighbor queries in SQL.

    Vector stores

Last check: . Method: public repository and docs read automatically, page and day written down. Next check due within 90 days. How we check

Is this your tool? Write to hello@agent-tools.directory from an address on its domain to correct a pin or a fact.