Committed usage for production AI — Agents API, Image API, and managed agent runs — with SLAs, priority routing through 50+ models, and cost controls that keep AI under 0.5% of your revenue.
Platform controls
Committed usage with the controls to match.
Spend analytics
Wallet burn, component costs, and token volume over time.
30-day spend trend
SLA uptime
99.9% uptime with service credits when we miss.
SLA: 99.9% uptime · credits on breach
Credit wallet
Prepaid balance with real-time burn rate tracking.
Balance
$0
Platform tiers
Builder, Scale, and Enterprise — better economics at every level.
Access management
Role-scoped members with per-agent API keys.
API health
All systems operational — status page in real time.
Three APIs, one contract
Every production pattern you need, billed together.
Managed Agents API
Billed per session / run
Embed streaming assistants and autonomous runs in your product — multi-tenant by design.
In-product copilots, customer-facing AI, B2B2C agent features.
Agents API
Billed per run
Autonomous workflow execution — run, cancel, observe, retrieve.
Background processing, nightly jobs, event-triggered automation, data pipelines.
Image API
Billed per operation
Gemini-backed generation, enhancement, and analysis with domain profiles.
E-commerce product imagery, medical imaging, document digitization, visual QA.
Platform tiers
Committed volume. Better economics at every level.
Annual contract. Unused runs roll over within the quarter. Enterprise tiers negotiated directly.
Builder
$30,000 committed annually
- 3,000 runs / month
- 500 image ops / month
- SLA: 99.5% uptime
- Support: 24-hour email
Production
$96,000 committed annually
- 10,000 runs / month
- 2,000 image ops / month
- SLA: 99.9% uptime
- Support: 4-hour email + Slack
Enterprise
Custom
- Unlimited runs
- Custom image volume
- SLA: 99.99% uptime
- Support: Dedicated success engineer
Cost efficiency
of revenue spent on AI
Across our deployments, AI cost runs under half a percent of the revenue it supports — versus 15–25% COGS for typical SaaS infrastructure. Caching, retrieval, and routing keep it there as you scale.
Prepaid credit wallet
Every turn is metered and debited in real time. A funded wallet gates spend before a single token is billed.
Budget caps & alerts
Monthly policy budgets with alerts at 50/80/100% — and an optional hard stop, even on a funded wallet.
Prompt + result caching
Static prefixes and tool schemas are cached, so multi-step turns don't re-pay for the same context.
Task-aware model router
Cheap models for classification and lookups; frontier models only when the task truly needs them.
Uptime and support
SLAs with credits. Support that actually responds.
Service credits apply when SLA is missed. Credit percentage scales with downtime duration. Enterprise SLAs include financial penalty clauses on request.
Start the conversation
Not sure which layer you need?
Most teams start with the Agents trial on Agents, prove ROI in month one, then embed via API when they're ready to embed. Platform contracts follow naturally when volume predictability matters. We can walk you through the math.