46.4kBook a DemoLog InGet Started Free

Zilliz Cloud Pricing

Built to scale with your usage, meet your security and compliance needs, and remain cost-effective for any budget. Estimate Cost

Subscribe on Marketplace.

AWSAWS
GCPGCP
AzureAzure

Free

Starting point for learning and personal projects.

$0
Start for Free
  • 5 GB storage
  • 2.5M
    vCUs
    per month included
  • Up to 5 collections
  • Community support: Slack invite link

Standard

Managed essentials for non-critical workloads. Best for: prototypes and testing environments.

From $0 /month (Serverless)
From $126/GB/month (Dedicated)
Start 30-day Free Trial
  • Fully managed vector databases with core APIs
  • Backup, restore, and basic monitoring
  • Built-in encryption for data in transit and at rest
  • Community support: Slack invite link

EnterpriseMOST POPULAR

Enterprise-grade reliability and controls. Best for: production applications.

From $197 /month (Dedicated)
Start 30-day Free Trial
  • 99.95% uptime SLA
  • Audit logs, SSO (SAML 2.0 based), granular RBAC
  • Multi-replica and elastic scaling
  • Private endpoint and VPC peering
  • On-demand
  • included

Business Critical

Regulated-ready with maximum resilience. Best for: healthcare, finance, and other highly regulated, mission-critical systems.

Let's Talk
Book a Demo
  • Global cluster with high-level availability and disaster recovery
  • Advanced security: CMEK and full-path in-transit encryption (edge TLS + internal)
  • HIPAA-eligible with enhanced data privacy features
  • and rapid incident response

On-demand Compute

New!

Run lake-scale search and indexing jobs on zero-copy external data.

  1. Zero-copy access to external data
  2. On-demand query jobs and system-managed index builds
  3. Pay only for active job runtime, with no always-on compute
Book a Demo

BYOC (Bring Your Own Cloud)

Designed for organizations that prioritize custom infrastructure, enhanced data protection, and compliance.

  1. Deploy on your infrastructure of choice
  2. High-level control and security
  3. Same features and experience as SaaS Dedicated clusters
Book a Demo
Compare Plans

Flexible Deployment Options

Dedicated

Dedicated provides isolated, reserved environments for production workloads that demand consistent and predictable performance. This option is ideal for sustained high-throughput and latency-sensitive applications.

  • Predictable Performance

    Leverage dedicated compute units (CU) to ensure stable performance without resource contention.

  • Fixed and Transparent Costs

    Easily manage your budget with a clear, pay-as-you-go pricing model.

  • Full Control & Customization

    Gain granular control over your resources and access advanced features to meet your specific application requirements.

Cluster Types

Dedicated clusters offer diverse types tailored to fit your use cases. A CU (compute unit) is a measure of computational resources for data processing. Each cluster type offers different combination of CPU, memory, and storage.

Performance-optimized

Ideal for real-time applications requiring instant search results and high concurrent traffic.

Per CU Capacity

2 million vectors

Search QPS

500-1500

Search Latency

10 ms

From $63

per million vectors / month

Capacity-optimized

Perfect for applications handling large vector datasets while maintaining reliable search speeds.

Per CU Capacity

8 million vectors

Search QPS

100-300

Search Latency

50-100 ms

From $16

per million vectors / month

Tiered-storage

Best for ultra-large-scale, cost-sensitive workloads.

Per CU Capacity

40 million vectors

Search QPS

10-50

Search Latency

100-1000ms

From $5

per million vectors / month

*This data is based on evaluations of 768-dimensional vectors.

View Plan Feature Details

Deployment
Deployment
Environment
SharedSharedDedicatedDedicatedDedicated
Cluster Type
Performance-optimized
Ideal for applications requiring low latency and high throughput. Each query CU can handle about 2 million 768-dim vectors.
Capacity-optimized
Suited for managing large datasets with moderate search performance requirements. Each query CU can handle about 8 million 768-dim vectors.
Tiered-storage
Best for ultra-large-scale, cost-sensitive workloads. Each query CU can handle about 40 million 768-dim vectors.
Public cloud provider
AWS, Google CloudAWS, Google CloudAWS, Google Cloud, AzureAWS, Google Cloud, AzureAWS, Google Cloud, Azure
Compute Scaling
System-managed auto-scaling
(No configuration required)		Manual scaling to
32 CUs		Configurable auto-scaling
Manual scaling up to 2,048 CUs; CU × replica ≤ 204,800
(Contact us for larger needs.)		Configurable auto-scaling
Manual scaling up to 2,048 CUs; CU × replica ≤ 204,800
(Contact us for larger needs.)
Uptime SLA
99.95%99.99% if multi-replica is enabled

Estimate Your Cost

Use this calculator to understand how Zilliz Cloud charges.

Cloud Provider
Region
Pricing Plan
Cluster Type
Number of Entities
An entity is a row of data. For example, an entity can represent an image or a doc chunk.
1
Million
Vector Dimensions
Vector dimension refers to the length of the vector and can vary depending on the embedding model you use.
768
Vector Data Type
mmap

Enabling mmap (memory mapping) can optimize memory usage and the amount of data that can be stored in the same Query CU will be increased. Learn More

Estimated total cost per month

$

Evaluate your cost with $200 free credits

Prices are estimates and may differ from actual costs. We recommend running a Proof of Concept (PoC) using the free credits to validate costs and performance, or contacting us for a tailored cost-optimization plan and additional PoC resources.

  • Query CU

    $

  • GB

    Storage

    $

Try FreeBook a Demo

Frequently Asked Questions

  • What is a Compute Unit (CU)?

    A compute unit (CU) is a group of hardware resources for serving your indexes and search requests. You can simply consider a CU as a fully-managed physical node for deploying search service.

  • What is a vCU?

    A virtual compute unit (vCU) is used to measure the resources consumed by read operations (such as search and query) and write operations (such as insert, upsert, and delete). The read and write costs vary for different vCU usage. The read and write costs vary for different vCU usage

  • Which type of cluster should I pick?

    Select the Performance-optimized cluster if you instant search results and high concurrent traffic for real-time applications.
    Choose the Capacity-optimized cluster if you need to handle large vector datasets while maintaining reliable search speeds.
    Opt for the Tiered-storage cluster if you need to handle ultra-large-scale, cost-sensitive workloads.

  • How many query CUs do I need for a given collection?

    Performance-optimized: Supports up to 2 million 768-dimensional vectors.

    Capacity-optimized: Supports up to 8 million 768-dimensional vectors.

    Tiered-storage: Supports up to 40 million 768-dimensional vectors.

    These estimates are based on vectors with primary keys only. Additional scalar fields like IDs or labels may reduce capacity. We recommend conducting your own tests for accurate assessment.

  • How can I get Zilliz Cloud discounts?

    When you commit to an annual plan, you can get additional credits based on your usage.

  • How can I request a new cloud region?

    To request a new cloud service provider region for Zilliz Cloud, please fill out the form.