Govern every AI request without freezing model choice.
Neural Router gives platform teams one controlled route to the model ecosystem, with policies for access, budgets, residency, retention, observability, provider eligibility, and service tiers.
One control plane for access and policy.
Manage orgs, workspaces, roles, API keys, management keys, BYOK, audit logs, privacy settings, and policy changes from a single governed surface.
- Orgs, workspaces, and roles
- API keys and scoped management keys
- Bring your own provider keys (BYOK)
- Audit logs on key and policy changes
- Privacy and residency settings
- Budget and rate-limit policies
Keep routing evidence close to your incidents.
Export route decisions and request telemetry into your observability stack, so routing evidence sits next to incidents, cost reviews, and compliance workflows.
See routing receiptsCapacity models for high-volume and sensitive traffic.
Priority
Requests receive higher priority and rate ceilings in shared capacity pools.
Platinum
Guaranteed baseline capacity with elastic priority burst for traffic spikes.
Dedicated
Reserved capacity for stronger isolation or predictable throughput. Dedicated-Isolated means no other customer traffic touches the reserved endpoints. Dedicated-Burstable allows preemptible spillover and credits idle capacity back to you.
Route sensitive traffic through eligible infrastructure.
Apply residency, retention, provider eligibility, and audit requirements so sensitive requests only reach endpoints that satisfy your envelope. Use routing receipts and evidence packs to support reviews. Formal certifications are in progress and available for procurement discussions.
Residency
Retention
Provider eligibility
Audit and evidence
Bring your platform team one governed route.
Talk to us about SSO, residency, dedicated capacity, custom SLAs, and procurement support.
Contact sales