[ RESOURCE_ALLOCATION ]

PRICING
MODELS

[ RESOURCE_ALLOCATION ]

PRICING_MODEL.CONF

// Select infrastructure tier based on active memory requirements and query throughput.

TIER_01

SANDBOX

$0/mo

Localhost development and prototyping.

MEMORY
1GB
LATENCY
Standard
RETENTION
7 Days
Included_Modules:
  • Public API Access
  • Shared Tenancy
  • Community Support
TIER_02

PRODUCTION

$299/mo

Scalable infrastructure for live agents.

MEMORY
100GB
LATENCY
Edge (<50ms)
RETENTION
Permanent
Included_Modules:
  • Managed Reranking
  • Private VPC Isolation
  • 99.9% Uptime SLA
TIER_03

HYPERSCALE

CUSTOM

Mission-critical global mesh.

MEMORY
∞
LATENCY
Edge (<20ms)
RETENTION
Permanent
Included_Modules:
  • Global Edge Routing
  • SSO & ACLs
  • Dedicated Solutions Eng.
[ KNOWLEDGE_BASE ]

FREQUENTLY ASKED QUESTIONS

How is 'Memory' calculated?

Memory usage is based on the total size of vector embeddings and metadata stored in your index. We compress data automatically.

What happens if I exceed my plan limits?

We offer a grace period of 24 hours. After that, read operations will continue, but write operations will be paused until you upgrade.

Can I deploy to my own VPC?

Yes, the Hyperscale plan supports private VPC deployments on AWS, GCP, and Azure with peering.

Do you offer a discount for open source projects?

Absolutely. Contact us with your repository details for a free Pro tier upgrade.

NEED CUSTOM INFRASTRUCTURE?

For mission-critical workloads requiring dedicated hardware, custom SLAs, and on-premise deployment options.