education
how pricing works
Transparent by design. Flat platform fee + usage at provider list prices. No lock-in, full portability.
engagement ladder
discovery sprint
1–2 weeksDeliverables: Architecture diagram, prioritized backlog, rollout plan with risk assessment.
Validates technical feasibility, scopes MVP, and sets expectations before build.
mvp build
2–6 weeksScope-dependent. Includes deployment, testing, documentation, and team training handoff.
Price varies by data integrations, compliance requirements, and custom logic complexity.
managed service
flat + usageIncludes: Platform operations, monitoring, on-call support, first quota of search queries and index storage.
Usage beyond included quota billed at provider list prices. See "What drives usage" below.
enterprise private ai
+ infrastructureIncludes: Dedicated infrastructure, priority support (1-hour SLA), compliance documentation, monthly optimization reviews.
Custom pricing for multi-region, high-volume, or specialized compliance needs.
what drives usage
We use a flat platform fee to cover managed operations, monitoring, and support. We include a clear quota for monthly usage. Beyond that, usage is billed at provider list prices for search, storage, models, and network. You can bring your own cloud accounts. Budgets are enforced with caps, alerts, and prepaid blocks.
queries & invocations
- • Search API calls (vector, hybrid, keyword)
- • Model inference requests
- • Tool and function executions
- • Workflow orchestration steps
storage
- • Vector index storage (per GB)
- • Document and artifact storage
- • Backup and snapshot retention
- • Log and audit trail storage
compute
- • Batch embedding jobs
- • Scheduled evaluation runs
- • Data transformation pipelines
- • Custom processing workflows
network & quality ops
- • Downstream API calls (egress)
- • Third-party integrations
- • Evaluation and testing runs
- • Incident response budget
example mvp bundle
monthly fee includes:
- Managed platform operations, monitoring, and on-call support
- First 10,000 search queries per month
- Up to 50GB index storage
- Basic monthly reports and dashboards
usage beyond included quota:
Billed at provider list prices for search, storage, models, and network. Provider can be swapped per client policy. No markup—we pass through provider pricing directly.
budget controls
prepaid blocks
Purchase usage credits in advance with volume discounts. Alerts when balance runs low.
per-workspace caps
Set spending limits per team, project, or environment. Hard caps prevent overruns.
monthly rollups
Detailed usage breakdowns by service, workspace, and time period. Export for chargeback.
live dashboards
Real-time cost tracking with forecasting. See spend trends and optimize before month-end.
portability & lock-in prevention
bring-your-own cloud keys
Use your own Google Cloud, AWS, or Azure accounts. We never hold your data hostage.
export data and indexes anytime
Full data export in standard formats. Indexes can be migrated to other platforms.
switch providers without rebuild
Model-agnostic architecture means you can swap Claude for OpenAI, Gemini for Llama, without code rewrites.
Questions about pricing or want a custom quote?