Weaviate

Weaviate is an open-source, AI-native vector database with a managed cloud service that lets engineering teams run semantic, keyword and hybrid search over their own data from one engine. It suits enterprises building RAG, agent memory and search applications that would otherwise stitch together a separate database, search engine and embedding service.

Weaviate is an open-source vector database, written in Go and released under the BSD-3-Clause licence, that stores data objects alongside their vector embeddings so applications can combine semantic similarity search with structured filtering in a single query. It is developed by an Amsterdam-headquartered company of the same name, founded in 2019 and co-founded by Bob van Luijt. Core capabilities include hybrid search that fuses BM25 keyword scoring with vector search, multi-tenancy, replication, vector compression and quantization, role-based access control, object time-to-live and retrieval-augmented generation with reranking, exposed over gRPC, REST and GraphQL APIs with official Python, JavaScript/TypeScript, Java, Go and C# clients. Modules connect to embedding and model providers including OpenAI, Cohere, Hugging Face and Google, and the project integrates with LangChain, LlamaIndex, CrewAI, Haystack, Semantic Kernel and DSPy. The GitHub repository had 16,812 stars when v1.39.5 shipped on 15 September 2026. Teams can self-host with Docker or Kubernetes, deploy through the AWS and GCP marketplaces or Snowpark Container Services, or buy Weaviate Cloud, whose Premium tier spans AWS, Google Cloud and Azure across roughly 40 regions, with hosted embeddings and a Query Agent offered as metered add-ons. Weaviate reports more than 20 million open-source downloads and lists Instabase, Akamai, Bosch, Cisco, Intuit and Booking among its users. It raised a $50 million Series B led by Index Ventures in April 2023, and the RICOH Innovation Fund made an undisclosed investment in March 2026. In the market it competes with Pinecone, Milvus, Qdrant, Chroma and Postgres with pgvector, and independent reviewers credit its hybrid search while flagging memory consumption and a steep GraphQL learning curve.

Category

Infrastructure & Cloud

Pricing

Freemium, Usage-based, Subscription

Target Market

CTOs, Enterprise Developers, Data Scientists, ML Engineers

Deployment

Open-source, Self-hosted, Multi-cloud

Founded

2019

Headquarters

Amsterdam, Netherlands

Customers

Thousands (vendor claim); 20M+ open-source downloads

• ✓
  Hybrid search

  Fuses BM25 keyword scoring with vector similarity in one query, improving retrieval accuracy for RAG when exact terms matter.

• ✓
  Multi-tenancy

  Isolates each customer's data as tenants inside one cluster; DocsBot reports running more than 50,000 tenants in a single cluster.

• ✓
  Vector compression and quantization

  Compresses stored vectors to reduce memory footprint and hosting cost, which matters because reviewers cite high RAM usage on large datasets.

• ✓
  Replication and role-based access control

  Replication provides fault tolerance and RBAC provides granular permissions, with data encrypted in transit and at rest on Weaviate Cloud.

• ✓
  Model-provider modules

  Vectorizer and generative modules connect to OpenAI, Cohere, Hugging Face and Google, so embeddings can be created without building a separate pipeline.

• ✓
  gRPC, REST and GraphQL APIs

  Three API surfaces with official Python, JavaScript/TypeScript, Java, Go and C# clients fit most enterprise application stacks.

• ✓
  Hosted embeddings and Query Agent

  Weaviate Cloud offers hosted embedding models from $0.025 per million tokens and a Query Agent with 1,000 free requests per month.

• •
  Enterprise RAG knowledge assistant

  Ground an LLM assistant in internal documents using hybrid search and reranking, so answers draw on retrieved passages rather than model memory.

• •
  Multi-tenant SaaS search and chat

  Serve search or chatbot answers to thousands of customers from one cluster with per-tenant isolation, as DocsBot does with 50,000+ tenants.

• •
  Billion-scale similarity search

  Run similarity search across very large embedding collections; Loti reports 9 billion vectors in production on Weaviate.

• •
  Unstructured document processing

  Index many unstructured document types for extraction and retrieval workflows, as Instabase does across more than 450 supported data types.

• •
  Agent memory

  Give AI agents a persistent store of context queried by meaning, wired in through the LangChain, LlamaIndex or CrewAI integrations.

Weaviate Cloud publishes list prices: an always-free tier, Flex from a $45 monthly minimum, and Premium from $400 per month on a prepaid contract. Usage is metered on stored vector dimensions (from $0.00465 per million on Flex to $0.002718 on Premium Dedicated), storage ($0.12 to $0.1505 per GiB) and backups, with data transfer currently free during a promotional period. Hosted embeddings cost $0.025 to $0.065 per million tokens and the Query Agent is $30 per month for 4,000 requests. SSO/SAML requires Premium and HIPAA requires Premium on AWS. The open-source database is free to self-host.

This page was written from 10 sources, 7 on domains other than weaviate.io.

1. 1.weaviate.io — pricingvendor
2. 2.weaviate.io — securityvendor
3. 3.weaviate.io — weaviate.iovendor
4. 4.docs.weaviate.io — deploy
5. 5.github.com — weaviate
6. 6.github.com — releases
7. 7.prnewswire.com — weaviate raises 50 million series b funding to meet soaring
8. 8.ricoh.com — 0616 1
9. 9.aws.amazon.com — prodview ng2dfhb4yjoic
10. 10.hn.algolia.com — search