Pricing for production agent infrastructure
Start with AgentCache Core, add Guardrails or Knowledge when needed, and use Execution Drift Guard as the fastest path into production trust.
Savings Capture Audit
Percentage of recovered budget retained as platform overhead.
Solo Pilot
Development & Research
- 3 Active Agent Slots
- 10K Requests/mo
- Standard Cache Only
Swarm Fleet
Scale your production swarm
- 100 Active Agent Slots
- 1M Requests/mo
- L2 Semantic Resonance
- Guardrails-ready
- Execution Drift Guard
Cognitive Cluster
Infinite memory federation
- Unlimited Agent Slots
- 10M Requests/mo
- Full Sector Resonance
- Advanced Drift Monitoring
- Custom Safety Anchors
Add-Ons
Expand your swarm capability
Enterprise Packages
Media edge, file acceleration, and ontology pilots for teams with domain-specific requirements.
Execution Drift Guard
For teams already running agents and needing proof before those workflows scale further.
- Up to 50K evaluated runs per month
- Shadow-mode drift evaluation and receipts
- Operator onboarding and weekly review
- 90-day paid pilot scope
Agent Storage Core
For teams that need agents to store, recall, and share artifacts on top of the JettyThunder storage backbone.
- Artifact storage orchestration and memory linkage
- Namespace isolation for pilot workloads
- Storage and egress metered separately
- Direct design-partner implementation support
Frequently Asked Questions
What counts as a request?
Any API call to AgentCache, whether it's a cache hit or miss. Cache hits are free from your LLM provider.
What happens if I exceed my plan limit?
Starter users should upgrade when they need more than 10K requests. Enterprise packages are available when 10M requests or bespoke infrastructure is not enough.
Can I cancel anytime?
Yes. Cancel anytime with no penalties. Your data is retained for 30 days.
Do you offer discounts for startups?
Yes. We can bundle startup discounts and design-partner pricing for teams validating Guardrails, Knowledge, or enterprise vertical packages.
