Weaviate
Weaviate is an open-source vector database that allows users to store data objects and vector embeddings from ML-models and scale to billio...
Last verified:
What is Weaviate?
Weaviate is an open-source AI-native vector database that stores both data objects and their vector embeddings, enabling semantic search, hybrid search combining vectors and keywords, and advanced retrieval-augmented generation (RAG) workflows. It supports agentic AI applications by powering intelligent agents with semantic insights from stored data. Designed for cloud-native scalability, Weaviate handles billion-scale workloads with features like dynamic indexing, vector compression, and high availability through replication and multi-tenancy.
Key features include seamless integration with multiple ML models for embeddings, language-agnostic SDKs in Python, Go, TypeScript, and JavaScript, plus GraphQL and REST APIs. It offers production-ready tools like auto-scaling clusters, baseline security with RBAC, and enterprise options such as SSO, HIPAA compliance, and SOC 2 certification. Weaviate Cloud provides fully managed deployments, while the open-source version allows self-hosting with customizable configurations.
Ideal for AI developers building search applications, trustworthy chat experiences, and agent workflows, Weaviate serves startups, scale-ups, and enterprises needing fast prototyping to production. Its ecosystem includes pre-built agents, managed embeddings service, and community resources like Weaviate Academy for learning AI-native development. With over 20 million downloads, it accelerates shipping AI features without managing infrastructure.
Weaviate pricing
Pricing model: Freemium
Free 14-day trial with Sandbox cluster including full core DB toolkit, hybrid search, compression, RBAC, and forum support. Flex plan: Pay-as-you-go from $45/month minimum, shared cloud, 99.5% uptime, email support; priced on vector dimensions (from $0.01668/1M), storage ($0.255/GiB), backups ($0.0264/GiB). Premium: Prepaid from $400/month, shared/dedicated options, 99.9-99.95% uptime, advanced support; lower rates like $0.0139/1M vector dims. Enterprise Dedicated: Contact sales, global CSPs (GCP/AWS/Azure), HIPAA, custom SLAs. Add-ons like embeddings $0.025-0.065/1M tokens; data transfer free promotional.
Weaviate pros
- Open-source vector database core
- Semantic and hybrid search capabilities
- Supports RAG workflows natively
- Agentic AI and workflow integration
- Billion-scale horizontal scalability
- Vector compression for cost efficiency
- Multi-tenancy for isolated workloads
- High availability with replication
- Language-agnostic SDKs available
- Seamless ML model integrations
- GraphQL and REST API support
- RBAC and enterprise security
- SOC 2 and HIPAA compliance options
- Managed Weaviate Cloud service
- Dynamic indexing for flexibility
- Pre-built database agents
- Global cloud provider support
- Community-driven innovation
Weaviate cons
- Pay-as-you-go can lead to unpredictable costs
- Minimum monthly fees on paid plans
- Free tier limited to 14-day trial
- Backup retention varies by plan
- SSO only on Premium and above
- No phone support on Flex plan
- Limited regions on shared cloud
- Data transfer fees possible post-promo
- Metrics endpoint not on lower plans
Frequently asked questions about Weaviate
What is a 'vector dimension' in Weaviate pricing?
A vector dimension refers to one element in a vector representing an embedding for an object. The total number is the vector's dimensionality. Pricing uses total vector dimensions calculated as number of objects times vector dimensionality per index times replication factor for high availability.
Why is pricing based on total vector dimensions?
Total vector dimensions indicate resources required like memory and compute. It provides a simple calculable metric over abstract units like CPU cycles, with additional billing for storage of objects and embeddings.
How does compression affect costs in Weaviate?
Compression reduces computational and memory resources for search without changing stored vector dimensions count. More aggressive methods like quantization offer discounted prices for vector dimensions relative to uncompressed vectors.
What are the differences between Free Trial, Flex, and Premium plans?
Free Trial is 14-day sandbox with forum support; Flex is pay-as-you-go shared cloud from $45/mo with email support; Premium is prepaid from $400/mo with SSO, better SLAs, and dedicated options. Features like HA and hybrid search are across all.
Can I move tenants to different storage tiers?
Moving tenants between hot, warm, and cold storage is available only on Weaviate Dedicated Cloud or self-hosted Weaviate, not on shared Flex or Sandbox plans.
Is Weaviate Cloud SOC2 compliant?
Yes, Weaviate Cloud maintains a SOC 2 report certifying controls for security, availability, confidentiality, and integrity of customer data, accessible via Trust Portal.
Is Weaviate Enterprise Cloud HIPAA compliant?
Yes, Enterprise Cloud on AWS meets HIPAA for secure PHI handling; support for Azure and GCP in development.
Do objects without embeddings incur storage costs?
Yes, billed for persistent and backup storage of objects and vectors. Objects without embeddings require less storage, thus lower costs.
What cloud providers and regions are supported?
GCP standard, AWS coming soon on Flex/Premium; Dedicated on GCP, AWS, Azure across all regions. Pricing varies by provider and region.
Is open-source Weaviate the same as Weaviate Cloud?
Weaviate Cloud uses the open-source project but adds managed service features like embedding service, query agent, data import, eliminating admin tasks for scale.