Pay for tokens (no markup), CPU/Memory (trivial), and a 10% platform fee. No per-seat fees, no idle charges. Same pricing in our cloud or your AWS account.
Every agent session runs in its own sandbox and you pay for exactly what it consumes.
Deploys into your AWS VPC. No credentials, code, or LLM calls leave it.
Every agent session runs in its own sandbox and you pay for exactly what it consumes.
Deploys into your AWS VPC. No credentials, code, or LLM calls leave it.
| Per-seat fees | $0 | $0 |
| Tokens | At cost, no markup | Billed by AWS |
| Platform fee | 10% of token cost | 10% of token cost |
| CPU | $0.142 / vCPU-hour | Billed by AWS |
| Memory | $0.024 / GB-hour | Billed by AWS |
| Control plane | Ellipsis cloud | Your AWS VPC |
| Agent sandboxes | Ellipsis cloud | Your AWS VPC |
| Feature set | Full | Full |
Start on Managed SaaS and move to your own AWS account later. Same agents, same configs.
Both offerings run the full platform. No feature gates, no tiers.
You don't need a support package to use Ellipsis. The platform is fully self-serve; these are for enterprises that want a direct line to our team.
| Slack support channel | |||
| Guided onboarding | |||
| Help building custom agents | |||
| Self guided BYOC | |||
| Help wiring BYOK (Anthropic or Bedrock) | |||
| Help with BYOC deployments | — | ||
| Security questionnaires and procurement | — | ||
| Priority feature requests | — | — | |
| First response time | 48 hours | 24 hours | 4 hours |
| Support engineer | Shared | Shared | Named |
Walk through the platform with our team and get your first agents deployed.