Billing & Rate Limits
BIE is priced by analysis sessions and the depth of intelligence produced. Rate limits protect you and the engine, they are strict but generous.
Plans
Four managed plans, from a free Developer tier to Enterprise. A complete breakdown, including agent access and burst limits, lives on the pricing page. The short version:
| Plan | Monthly sessions | Max events / session | Agent access |
|---|---|---|---|
| Developer (free) | 100 | 2,000 | None |
| Starter | 1,000 | 5,000 | Cyrano |
| Growth | 5,000 | 10,000 | Cyrano, Argus |
| Scale | 25,000 | 10,000 | All agents |
| Enterprise | Custom | 10,000 | All agents + custom domains |
Rate limits
Every API response includes standard rate limit headers so you can back off before hitting a 429. The default window is one hour.
| Header | Meaning |
|---|---|
| X-RateLimit-Limit | Maximum requests allowed in the current window |
| X-RateLimit-Remaining | Requests left before a 429 |
| X-RateLimit-Reset | Unix timestamp when the window resets |
| Retry-After | (429 only) seconds to wait before retrying |
Defaults are 1000 requests per hour and 10000 per day per key. Drift analysis has its own stricter limit of 10 sessions per hour. Plans scale these limits proportionally.
Budget enforcement
Each key has a monthly budget tied to its plan. If usage exceeds the budget or spikes 5x above the rolling baseline, the key is suspended and further requests return 429 RATE_LIMITED. You can lift the suspension or raise the budget in the Developer Console.
Cost tracking
Every session records token usage and the AI cost it generated. You can see a per-session breakdown in the Developer Console under Recent Sessions, or pull it programmatically from the session detail endpoint.
TODO: expose a dedicated GET /usage endpoint with period summaries.