Transparent pricing up to Organization M, custom above. Every number below is labeled the way GAIA labels its own outputs: pilot-measured or estimated — never invented.
One business unit of 30–50 people, one workflow, 30 days
credited against your contract if you sign
100–250 employees
≈ €63 per seat, all included · estimated
250–500 employees
scoped to your stack and policies
500+ employees, or regulated / sensitive environments
private or hybrid deployment
Evaluating the API first? The developer sandbox is free — limited requests, clearly labeled sandbox keys, proof statuses on every output.
The economics
Illustrative model for a 180-person organization, from our first deployment scoping. Your pilot report replaces these with your measured baseline.
| Current stack (fragmented) | GAIA, year 1 | |
|---|---|---|
| Monthly AI spend | ~$24,200 | ~$13,700 |
| Annual comparable | ~$290k | ~$164k recurring + $11k onboarding |
| Per employee / month | ~$134 | ~$63 — all included |
| Cash ROI | — | Positive from month 8 |
Cost explainer
GAIA routes work across model tiers instead of sending everything to the most expensive model. Your costs depend on what your workflows actually consume:
Fast classifiers cost a fraction of frontier models. Routing puts each request on the cheapest tier that can do the job.
Input + output volume per request, after context-pack budgeting trims what enters the prompt.
Tenant-scoped recall operations and the storage behind them.
Governed tool runs — file operations, integrations, verifications.
How long proofs, traces and memory are retained for audit.
Dedicated infrastructure, isolation guarantees and security review scope.
One workflow in, one proof-backed cost/latency report out. That’s how pricing should start.