Customers
Built for teams putting AI into production.
Our design-partner program is open. See who nRouter is built for and how to become one of our first customer stories.
Four kinds of teams build on nRouter.
One key, one bill, and exact list-price settlement across production workloads.
Developer platforms and agent builders.
Route multi-step agent calls across models with automatic fallbacks and one unified API key.
- Your own fallbacks, tried when a provider cannot take the call
- Pay as you go from $5 with a 0% platform fee
- Usage settled at the served model's exact list price
Regulated and compliance-focused teams.
Keep request-level audit trails, strict tenant isolation, and predictable list-price billing.
- Tenant-isolated organization and team boundaries
- Request logs with model, routing path, and exact cost
- 0% platform fee and zero markup on model list prices
Multi-tenant SaaS and platform teams.
Scope virtual keys, budgets, and rate limits per team or workflow so one workload cannot overspend.
- Per-key model allowlists, route ACLs, and spend ceilings
- Pre-call credit reservation before upstream dispatch
- Refused requests before dispatch hold and spend $0
Privacy and safety-sensitive workloads.
Enforce pre-call and post-call guardrails for PII redaction, moderation, and prompt-injection defense.
- Scoped guardrails at the key, team, or organization level
- PII redaction and content moderation in the request path
- Blocked pre-call requests return an error at $0 cost
Be one of our first customer stories.
Join the design-partner program and shape what ships next.
A direct line to the team
Collaborate in a shared channel with the engineers building the gateway.Apply to become a design partnerShape the roadmap
Influence what ships next across routing strategies, providers, and dashboard surfaces.Apply to become a design partnerHands-on onboarding
Migrate from direct provider SDKs with an architecture walkthrough from our team.Apply to become a design partnerA case study, if you want one
Measure real production impact together and publish only with your written approval.Apply to become a design partner
What our first case studies measure.
Every published story tracks cost, reliability, and governance against real usage.
Cost saved.
0% platform fee across 371+ models and spend avoided by routing each request to the right-priced model.
0%
Latency and reliability.
Response latency and completion success rate held up by pre-stream fallback chains when a provider degrades.
99.9%
Governance and safety.
Guardrail catches, budget breaches prevented, and request-level audit trails for security reviews.
$0
FAQs
How do you validate performance claims and case studies?
We validate latency, throughput, and cost reduction across high-concurrency workloads and design-partner production runs. Where partners are in launch stages or under pilot agreements, architecture patterns and metrics reflect measured benchmark results from those deployments.
What does the design-partner program involve?
A direct channel to the engineering team, influence over the roadmap, hands-on migration support, and direct onboarding. In return we ask for candid feedback and, optionally, permission to write up the results once they are real.
How fast can we be in production?
Self-serve onboarding takes minutes — sign up, create a virtual key, point your base URL at nRouter, and call any model. Design partners get a hands-on onboarding session and a migration walkthrough from their current provider setup.