Free
$0per month
The hosted gateway, free. Point your agents at one endpoint and route through your own provider keys and local models.
- Every provider, one endpoint
- Your own keys and local models
- Budgets, allowlists, attribution
- Community support
Pricing
Start free, pay providers exactly what they charge, and add the intelligence layer when you want it.
Pro
0%markup on tokens
Route on credits at each provider’s list price, with nothing on top. One bill across every provider.
- Everything in Free
- Credits and top-ups
- One bill across all providers
- Mix credits and your own keys freely
- OpenAI-compatible, streaming included
- Every provider behind one endpoint
- Your own keys and local models
- Budgets that stop requests at the cap
- Every token attributed
- Open source core
| Free | Pro | |
|---|---|---|
| Gateway | ||
| OpenAI-compatible /v1 endpoint | Included | Included |
| Every hosted provider | Included | Included |
| Your own provider keys | Included | Included |
| Local and custom models | Included | Included |
| Provider failover | Included | Included |
| Markup on tokens | 0% | 0% |
| Credits and top-ups | · | Included |
| One bill across providers | · | Included |
| Controls | ||
| Budgets and hard caps | Included | Included |
| API key management | Included | Included |
| Model allowlists per key or agent | Included | Included |
| Dashboard | Included | Included |
| Observability | ||
| Usage by agent, person, model, day | Included | Included |
| Request logs with route and cost | Included | Included |
| Live traffic view | Included | Included |
| Security | ||
| Route only to ZDR providers | Included | Included |
| No-training provider policy | Included | Included |
| Provider allowlists | Included | Included |
| Provider keys never reach agents | Included | Included |
| Support | ||
| Community (Discord and GitHub) | Included | Included |
Routing policies, budgets, and allowlists are enforced at the gateway on every request. No code changes.
Enterprise-ready.
The posture an enterprise rollout asks for, labeled honestly: available today, on request, or on the roadmap. For the intelligence layer, security reviews, and dedicated support, talk to us.
Multiple organizations
Separate orgs, keys, and billing under one account.
Google and GitHub sign-in
OAuth sign-in for the whole team, no passwords required.
Usage history and attribution
Usage, spend, requests, TTFT, and token counts at key and team scope.
Usage API
Spend by agent, person, model, or day, pulled into your own systems.
API key management
Create, view, and delete keys from the dashboard or the API.
Zero Data Retention routing
Route only to providers under a ZDR agreement, enforced in policy.
No training on your data
Route only to providers that will not train on customer data.
Provider resilience
Failover across providers and pooled accounts on capacity errors.
Request logs
Every request recorded with its route, tokens, and cost.
Budgets and policies
Hard caps and model allowlists, enforced on every request.
SOC 1
SOC 1 compliance, report available on request.
SSO: SAML and SCIMContact us
Enterprise identity providers and directory sync.
Short-lived credentialsContact us
Expiring tokens in place of static keys.
Advanced RBACContact us
Fine-grained roles beyond the built-in org roles.
Org-wide policy packsContact us
One policy set applied across every team and key.
Approval workflowsContact us
Changes reviewed before they land, with change history.
Managed KMSContact us
Customer-managed keys for stored provider credentials.
Private networkingContact us
Private connectivity between your VPC and the gateway.
Data residencyContact us
Regional routing and storage constraints.
Enterprise supportContact us
A named contact and a shared channel.
Managed upgrades and backupsPlanned
For self-hosted deployments we operate with you.
Managed optimization workersPlanned
Dedicated capacity for the intelligence layer.
Forecasting and alertsPlanned
Spend projections and threshold notifications.
Questions
Is there really no markup?
Yes. Tokens cost the provider's list price whether you route on credits or on your own keys. We add $0.00 per token; the paid product is the intelligence layer, not a percentage of your traffic.
What happens when my credits run out?
Requests on credits stop at your caps, so nothing silently overruns. Top up from the dashboard, and anything routed through your own provider keys keeps working the whole time.
Can I bring my own provider keys?
Yes, free, on every plan. The gateway routes through your keys first, so existing provider commitments and discounts flow through untouched.
What does self-hosting include?
The whole gateway, not a demo: the OpenAI-compatible /v1 endpoint, key management, budgets, provider waterfalls, and the usage API. uvx --from experiential exp run starts it on your own infrastructure.
Route your first request today.
One base URL swap. Provider prices, hard budgets, every token attributed.