Weaviate
Free tierThe open-source AI-native vector database built for production-scale search, RAG, and agents
Free tier available·All audiences·API available·Open source
Key strengths
Billion-scale vector search with multi-tenancy supportBuilt-in embeddings — no external pipeline requiredDeployment-agnostic: cloud, self-hosted, or on-premUnified platform for search, RAG, agents, and memory (Engram)Production-hardened with SOC 2, HIPAA, RBAC, and high availability
Free tier + paid plans
Amsterdam, Netherlands
Founded 2019
Self-hostable
No ratings yet
Developer Documentation
Weaviate offers comprehensive documentation across multiple surfaces — database, cloud, deployment, and agents — all available at weaviate.io/developers.
Key developer resources:
- SDKs: First-class clients for Python, Go, TypeScript, and JavaScript, with idiomatic collection-based APIs.
- APIs: Full GraphQL and REST API support for flexible querying and schema management.
- Search modes: Pure vector (
near_vector), semantic (near_text), and hybrid (vector + keyword with tunablealpha) search out of the box. - Multi-tenancy: Native tenant isolation allowing 50K+ segmented indexes within a single cluster — ideal for SaaS workloads.
- Modules & integrations: Plug-in vectorizer modules (OpenAI, Cohere, HuggingFace, etc.) and integrations with AWS, Google Cloud, Snowflake, and Databricks.
- Agents: Weaviate Agents Docs cover the Query Agent and upcoming agentic AI capabilities; LLM/Agent guides provide integration patterns.
- Observability: Built-in observability, RBAC, SOC 2 & HIPAA compliance, and high-availability configurations for enterprise deployments.