Zilliz
Zilliz is the commercial company behind Milvus, the leading open-source vector database. It provides managed vector data infrastructure (Zilliz Cloud / Vector Lakebase) to enterprises building AI applications such as RAG, semantic search, and AI agent memory, serving 10,000+ enterprise teams globally.
- Company typePrivate
- Founded2016
- HeadquartersRedwood City, United States
- Headcount101–250
- GTM typeB2B
- OfferingSoftware
What Zilliz does
Zilliz is the commercial company behind Milvus, the leading open-source vector database, serving enterprises building AI-native applications such as retrieval-augmented generation, semantic search, recommendation systems, and AI agent memory. Founded in 2017 and headquartered in Redwood City, California (relocated from China in 2022), the company combines an Apache 2.0 open-source core (Milvus, with 44,900+ GitHub stars and adoption by 10,000+ enterprise teams including xAI, NVIDIA, Zillow, Reddit, ByteDance, Bosch, Airtable, and Roblox) with a managed commercial product, now branded as Zilliz Cloud and repositioned as Vector Lakebase — a unified vector data platform spanning storage, indexing, and retrieval.
The technology stack layers proprietary IP on top of the open-source core, including the Cardinal Search Engine with AI-powered AutoIndex, the Loon storage engine, the Vortex columnar format, and Global Cluster / Cross-Region Disaster Recovery for enterprise resilience. Zilliz delivers its managed service through three deployment models — fully managed, BYOC (Bring Your Own Cloud) generally available across AWS, Azure, and GCP, and self-hosted — with presence on all major cloud marketplaces. The company meets enterprise compliance requirements with SOC 2, ISO 27001, HIPAA, PCI DSS, and CMEK certifications.
The business model pairs an open-source land-and-expand motion (free community edition driving developer adoption) with tiered paid commercial offerings, enterprise contracts, and hyperscaler co-sell relationships including a strategic partnership with AWS announced in December 2025. As of December 2024 the company employed approximately 136 people; total disclosed funding is $113M including a $60M Series B extension led by Prosperity7 Ventures in August 2022, and reported 2023 revenue was $16.2M.
Zilliz firmographics
Firmographics- Name
- Zilliz
- Legal name
- Zilliz Inc.
- Website
- https://zilliz.com
- Company type
- Private
- Founded year
- 2016
- Operating status
- Operating
- Headcount range
- 101–250 employees
- Short description
- Zilliz is the commercial company behind Milvus, the leading open-source vector database. It provides managed vector data infrastructure (Zilliz Cloud / Vector Lakebase) to enterprises building AI applications such as RAG, semantic search, and AI agent memory, serving 10,000+ enterprise teams globally.
- Ownership category
- akta.pro rank
Where Zilliz is headquartered
LocationHeadquarters
- HQ city
- Redwood City
- HQ country
- United States
- HQ region
- North America
Offices2 records
Markets served
Zilliz business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations
Revenue model
- Zilliz Cloud SaaS subscriptions: Recurring SaaS revenue across Free, Standard, Enterprise, and Business Critical plans with monthly billing and free trial credits ($100–$200). Plans available on AWS, GCP, and Azure marketplaces.
- Usage-based compute and storage: Compute billed by CU-hour ($0.175/CU-hour for performance- and capacity-optimized; $0.263 for tiered-storage) plus storage at $0.025/GB/month (rising to $0.040 in Jan 2026), with pay-as-you-go vCU consumption for reads/writes and on-demand CU-minute billing for On-Demand Search.
- Freemium Free Tier: Free tier ($0/month) with 5 GB storage and 2.5M vCUs/month drives adoption and conversion to paid Standard, Enterprise, and Business Critical tiers.
- Bring Your Own Cloud (BYOC) contracts: Pricing based on CPU core consumption of data services requiring a minimum annual commitment; customers can apply existing cloud credits and discounts on top of Zilliz Cloud infrastructure spend.
- Business Critical / enterprise contracts: Quote-based, contact-sales revenue for the Business Critical plan (HIPAA, PCI DSS, CMEK, Global Cluster, 99.99% uptime SLA) tailored to regulated industries and mission-critical deployments.
- AWS Marketplace and strategic cloud partner revenue: Listed on AWS Marketplace, GCP Marketplace, and Azure Marketplace; AWS re:Invent 2025 partnership announcement deepened distribution to hundreds of enterprise customers.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Monthly | Free Tier |
| Hybrid | Monthly | Standard |
| Subscription | Monthly | Enterprise (Dedicated) |
| Subscription | Multi-year contract | Business Critical |
| Usage-based | Pay-as-you-go | On-Demand Compute |
| Subscription | Multi-year contract | BYOC (Bring Your Own Cloud) |
| Unit Pricing | Monthly | Dedicated Cluster Types |
| Other | Annual | AI Startup Program |
Go-to-market motion1 record
Distribution channels7 records
Marketing channels12 records
Zilliz product offering
Product offeringCore offering
Zilliz provides a fully managed cloud-native vector database and Vector Lakebase platform built on its open-source Milvus vector database, enabling enterprises to store, index, and search billion-scale unstructured data through high-dimensional vector embeddings. The platform supports retrieval-augmented generation (RAG), semantic search, recommendation systems, and agentic AI workloads across Serverless, Dedicated, and Bring-Your-Own-Cloud deployment models.
Product overview
Zilliz delivers a platform-plus-modules AI data architecture centered on its open-source Milvus vector database and the fully managed Zilliz Cloud service, now rebranded as the Vector Lakebase for AI. The core engine is Milvus, the Apache 2.0 open-source vector database built for billion-scale similarity search, which is paired with the Cardinal Search Engine and Loon storage engine in the managed cloud. Zilliz Cloud itself is offered in multiple deployment options (Dedicated, Serverless, BYOC) and tiered plans (Free, Standard, Enterprise, Business Critical), with the new Vector Lakebase layer adding lake-native storage on the Vortex open columnar format to unify real-time serving, iterative discovery, and batch analytics. Surrounding this core are a suite of complementary products and tools: Zilliz Migration Service (and its open-source sibling Vector Transport Service / VTS) for moving data between systems; PyMilvus and Knowhere as the SDK and search engine; Milvus Backup, Milvus CLI, Milvus Sizing Tool, Attu, and VectorDBBench for operational tasks; and AI-focused add-ons including GPTCache, DeepSearcher, Claude Context, Feder, the Bilingual Semantic Highlighting Model, and memsearch for AI agent persistent memory. Together these offerings span the full lifecycle from ingestion and embedding through retrieval, RAG, and agent deployment.
Differentiator
Problem solved
Functional benefit
Brands
- Zilliz Cloud: Fully managed vector database / Vector Lakebase service built on Milvus.
- Milvus
- Vector Lakebase
- Cardinal Search Engine
- GPTCache
- DeepSearcher
- Claude Context
- Attu
- VectorDBBench
- VTS (Vector Transport Service)
Products and services
- Zilliz Cloud A fully managed cloud-native vector database and data services platform built on Milvus, designed to help enterprises unlock unstructured data for AI applications. Available in Serverless, Dedicated, and BYOC deployment models with tiered plans from Free to Business Critical.
- Zilliz Vector Lakebase A unified AI data platform that pairs production vector search with a shared lake-native data foundation, enabling real-time serving, interactive discovery, and batch analytics on a single logical copy of data, scaled from gigabytes to petabytes. Built on the Vortex open columnar format and the Loon storage engine.
- Milvus An open-source vector database under Apache 2.0 license that can store, index, and search billion-scale unstructured data through high-dimensional vector embeddings. Used for RAG, semantic search, multimodal search, and recommendation systems.
- Zilliz Cloud BYOC (Bring Your Own Cloud) A deployment option that runs Zilliz Cloud within the customer's own VPC on AWS, GCP, or Azure, separating control plane from data plane for complete data sovereignty. Zilliz manages provisioning, scaling, and maintenance while customer data never leaves the customer's cloud environment.
- Zilliz Migration Service A free, fully managed migration tool for unstructured and vector data with zero-downtime migration and offline batch import from sources including Elasticsearch, OpenSearch, Pinecone, Qdrant, Milvus, and PostgreSQL. Built on the open-source Vector Transport Service (VTS).
- Zilliz Cloud Free Tier Free tier of Zilliz Cloud with 5 GB storage, 2.5M vCUs per month, and up to 5 collections, designed for learning, prototyping, and proof of concept. Includes two collections housing 1,000,000 768-dimensional vectors and built-in vector embedding.
- Zilliz Cloud Business Critical Plan Premium service tier in Zilliz Cloud for mission-critical AI workloads. Includes Global Cluster, multi-region replication, automated failover, PITR, CMEK, full-path in-transit encryption, HIPAA and PCI DSS compliance, and 30-minute emergency SLA.
Quantifiable outcome
- Sub-10ms latency at billion-scale vector search on Zilliz Cloud
- +10 more outcomes
Companies that use Zilliz
Customer profileNamed customers27 records
Segments6 records
Ideal customer profiles5 records
Zilliz technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration28 records
AI capability14 records
Feature12 records
Zilliz partnerships and signals
Strategic signalPartnerships
17 partnerships are on record, tiered flagship and core.
- Microsoft AzureflagshipBYOC on Microsoft Azure launched in GA, completing rollout across all three major cloud platforms (AWS, GCP, Azure). Makes Zilliz the first managed vector database provider with BYOC on all three major clouds; integrates with Azure OpenAI Service.
- KIOXIA AmericacoreKIOXIA's AiSAQ approximate nearest neighbor search (ANNS) software integrated into Milvus starting with version 2.6.4, enabling dramatic DRAM reduction for large-scale Retrieval Augmented Generation deployments and supporting trillions of vectors on SSDs.
- Amazon Web Services (AWS)flagshipAWS selected as Zilliz's strategic cloud provider. Zilliz deployed across nine AWS Regions spanning North America, Europe, and Asia, with over 20% cost optimization via Amazon Graviton. Zilliz Cloud integrates with Amazon Bedrock (foundation models), Amazon SageMaker, and Amazon Q; new region deployment time reduced to one week (down from one month).
- PliopscoreCollaboration to enable billion-scale vector search at storage-level costs by combining Pliops' LightningAI architecture with Milvus software. Targets multi-billion-scale vector databases via hardware-accelerated KV-Cache Offload, multi-tier storage, KV mapping, and dual-tier flash/S3 architecture. Demonstrated live at Supercomputing 2025 (SC25).
- CloudiancoreCloudian integrated Milvus vector database into its HyperStore object storage software to provide AI inference as part of its AI Data Platform roadmap. The combined platform eliminates data movement between separate unstructured and vector stores, supports on-premises and hybrid cloud deployments, and is available for evaluation.
- Amazon BedrockcoreJoint deployment of enterprise-grade vector search with Amazon Bedrock and Zilliz Cloud for managing unstructured data and enhancing large language models with real-time, contextual information.
- Google CloudflagshipZilliz Cloud built on Google Cloud infrastructure using Kubernetes and microservices; achieved 99.95% SLA reliability, scaled to thousands of clusters, and serves over 10,000 global enterprise customers as of 2024. Available on Google Cloud Marketplace with three operations employees managing the platform.
- OpenAI (ChatGPT Retrieval Plugin, text-embedding-3 series, CLIP)coreOfficial Zilliz Cloud integration with OpenAI ChatGPT Retrieval Plugin and OpenAI embedding models (text-embedding-3-large, text-embedding-3-small, ada-002) plus CLIP-ViT-B/32 multimodal model, supporting RAG, semantic search, and multimodal AI.
- LangChaincoreOfficial Zilliz Cloud integration with LangChain for building LLM applications with semantic search, RAG, and context-aware retrieval using a flexible orchestration framework.
- CoherecoreIntegration with Cohere's multilingual embedding models, NLP capabilities, and Cohere Reranker for high-precision cross-encoder reranking in Vector Lakebase.
- Voyage AIcoreIntegration with Voyage AI embedding models (voyage-2, voyage-large-2, voyage-code-2) and Voyage AI Reranker for fast, cost-efficient relevance scoring in large-scale retrieval pipelines.
- Jina AIcoreIntegration with Jina AI embedding models (jina-embeddings-v2-base-en, v2-small-en) and rerankers for multilingual, multimodal, and code similarity search at scale.
- DatabrickscoreReal-time ingestion of vectors from unstructured data via Databricks connector.
- Confluent (Apache Kafka)coreReal-time data ingestion for RAG applications via Confluent/Kafka connector.
- DatadogcoreDatadog integration for monitoring, visualizing, and optimizing vector database performance in AI-powered applications.
- LlamaIndexcoreOfficial Zilliz Cloud / Milvus integration with LlamaIndex for building powerful RAG systems with enhanced text generation.
- vLLMcoreIntegration with vLLM's optimized LLM inference (PagedAttention) to deliver up to 24x throughput improvement alongside Zilliz Cloud's scalable vector database for RAG systems.
Scale indicators17 records
Recent moves10 records
Expansion highlights6 records
Zilliz competitors and assessment
Company assessmentDirect peers
- Pinecone: Pinecone is a venture-backed managed vector database that competes head-to-head with Zilliz Cloud in the serverless, production-grade vector search category. Both target RAG, semantic search, and recommendation use cases on AWS, GCP, and Azure, and both invest heavily in enterprise security, BYOC-like options, and cost-efficiency benchmarks — making them the most direct commercial comparables in the space.
- Weaviate: Weaviate is an open-source vector database with a managed cloud offering, similar in positioning to Milvus + Zilliz Cloud. Both are built around modular search (vector + hybrid + filtered), open-source community distribution, and enterprise deployments in RAG and semantic search, making Weaviate a structural twin for benchmarking open-source-driven vector database business models.
- Qdrant: Qdrant is an open-source vector database written in Rust with a managed cloud offering, directly comparable to Milvus in technical positioning and Zilliz Cloud in commercial model. Both target high-performance vector search, hybrid filtering, and enterprise RAG workloads, and both compete for the same developer-led growth channel.
Emerging players
- Chroma: Chroma is an emerging open-source embedding database popular with AI/ML developers and LLM application builders. It overlaps with Milvus/Zilliz Cloud at the developer and small-team tier, particularly in RAG and prototype workloads, and is increasingly being positioned as a competitor in the broader vector database conversation.
- LanceDB: LanceDB is an open-source vector database built on the Lance columnar format, conceptually adjacent to Zilliz's Vortex-based Lakebase. Both target developer-friendly multimodal AI workloads with embedded and serverless options, and both are pushing lake-native storage as a vector database differentiator.
- Vespa.ai: Vespa is a Yahoo!-originated open-source platform for low-latency computation over large data sets, including vector and hybrid search at scale. It overlaps with Zilliz in production-grade vector and hybrid retrieval for large enterprise workloads, particularly in recommendation and relevance scenarios.
Broad incumbents
- Elastic (Elasticsearch / OpenSearch): Elasticsearch and its open-source fork OpenSearch offer vector search and hybrid retrieval as part of a broader search/observability platform, making Elastic a direct functional competitor to Zilliz in hybrid search and full-text + vector workloads. Zilliz benchmarks itself against Elasticsearch (7x faster full-text search), underscoring the head-to-head nature of the comparison.
- MongoDB (Atlas Vector Search): MongoDB Atlas Vector Search embeds vector search directly into the Atlas document database, positioning MongoDB as a broad incumbent that can offer vector retrieval as part of an existing operational data store. For customers already on Atlas, this creates a strong bundled alternative to a separate Zilliz deployment.
- Google Cloud (Vertex AI Vector Search): Google Cloud's Vertex AI Vector Search (formerly Matching Engine) is a managed vector database offered as part of the broader Vertex AI platform. It competes with Zilliz Cloud for AI workloads on GCP and is one of the three hyperscaler offerings Zilliz's BYOC and marketplace strategy is designed to coexist with — and defend against.
- Databricks: Databricks offers vector search as part of its lakehouse and Mosaic AI stack, and has a partnership with Zilliz for real-time vector ingestion. As Databricks continues to push into AI infrastructure and retrieval, it represents both a partner and a potential long-term competitor to Zilliz's Vector Lakebase positioning.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat7 records
Key risks5 records
Key highlights7 records
Customer concentration
Zilliz social profiles
Digital presenceZilliz compliance and trust
Trust signalCompliance5 records
Zilliz financial estimates
Financial estimateRevenue estimate
Valuation estimate
Zilliz leadership team
Management profileNumber of profiles
Profiles7 records
Zilliz funding detail
Funding detailFunding overview
Funding rounds4 records
Investors9 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Zilliz M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Zilliz
What does Zilliz do?
Zilliz provides a fully managed cloud-native vector database and Vector Lakebase platform built on its open-source Milvus vector database, enabling enterprises to store, index, and search billion-scale unstructured data through high-dimensional vector embeddings. The platform supports retrieval-augmented generation (RAG), semantic search, recommendation systems, and agentic AI workloads across Serverless, Dedicated, and Bring-Your-Own-Cloud deployment models.
Is Zilliz a public or private company?
Zilliz is a private company. It is classified as venture growth investor backed and is currently operating.
When was Zilliz founded?
Zilliz was founded in 2016. It employs 101 to 250 people.
Where is Zilliz based?
Zilliz is headquartered in Redwood City, United States, in the North America region.
How does Zilliz make money?
Six revenue lines are on record. Zilliz Cloud SaaS subscriptions are the primary driver. The others are usage-based compute and storage, freemium Free Tier, bring Your Own Cloud (BYOC) contracts, business Critical / enterprise contracts and AWS Marketplace and strategic cloud partner revenue.
Who are Zilliz's main competitors?
Direct peers on record are Pinecone, Weaviate and Qdrant. Emerging players are Chroma, LanceDB and Vespa.ai. Broad incumbents are Elastic (Elasticsearch / OpenSearch), MongoDB (Atlas Vector Search), Google Cloud (Vertex AI Vector Search) and Databricks.
Does Zilliz have an API?
Yes. Zilliz Cloud offers intuitive RESTful APIs for control and data plane operations. SDKs are provided in Python, Java, Go, and Node.js. Zilliz Cloud integrates with Amazon Bedrock through APIs to access foundation models and supports model customization through Amazon SageMaker. The Milvus API supports operations including insert, search, query, vector search, filtered search, range search, grouping search, hybrid search, full-text search, text match, and data processing. Auto-scaling and elastic scaling are managed via API. Developer documentation is at docs.zilliz.com/docs/home.