Pay for what the models actually cost, plus a margin we publish.
One credit is one US cent of model value. Your plan includes a monthly allowance; if a month runs heavier, move up a plan. Endpoints and seats are counted, not metered.
Free
Two machines, real answers
Enough to investigate a machine you own on the fast tier.
2 seats
- 2 endpoints
- 200 credits a month
- Fast models
- Full read-only tool set
Starter
For the person who owns the estate
A small fleet, the balanced models, and history that outlives the incident.
5 seats
- 15 endpoints
- 5,000 credits a month
- Balanced models
- 2 concurrent investigations
- 2 API keys
Team
For a security function
Frontier models, real concurrency, exportable audit trail.
20 seats
- 100 endpoints
- 25,000 credits a month
- Frontier models
- 5 concurrent investigations
- Audit-log export
Enterprise
For the whole estate
For a large estate: the highest limits, and terms agreed with you.
500 seats
- 5,000 endpoints
- 150,000 credits a month
- Every enabled model tier
- 20 concurrent investigations
- 50 API keys
- Audit-log export
or write to sales@guardyn.dev
The arithmetic, in the open.
A credit is a cent of retail model value. Each model carries its own price and its own margin, both visible to whoever administers the platform.
| Model | Provider | Tier | Input, per million tokens | Output, per million tokens | Available on |
|---|---|---|---|---|---|
| Claude Haiku 4.5 | Anthropic | Fast | 300 cr | 1,500 cr | Free and up |
| Claude Sonnet 5 | Anthropic | Balanced | 600 cr | 3,000 cr | Starter and up |
| Claude Opus 5 | Anthropic | Frontier | 1,500 cr | 7,500 cr | Team and up |
Credits per million tokens
- Claude Haiku 4.5FastAnthropic
- Input
- 300 cr
- Output
- 1,500 cr
- Available on
- Free and up
- Claude Sonnet 5BalancedAnthropic
- Input
- 600 cr
- Output
- 3,000 cr
- Available on
- Starter and up
- Claude Opus 5FrontierAnthropic
- Input
- 1,500 cr
- Output
- 7,500 cr
- Available on
- Team and up
Credits are charged at the provider's list price × 3; one credit is US$0.01.
A typical single-question investigation on the balanced tier — four or five tool calls and a couple of pages of answer — lands between 8 and 25 credits. A full compromise assessment on the frontier tier is usually 60 to 200. Prompt caching means follow-up questions in the same conversation cost a fraction of the first.