nRouter.ai
What is nRouter.ai?
NRouter.ai is a managed LLM routing platform for product and platform teams that sits between your app and model providers, exposing an OpenAI-compatible endpoint so existing SDKs can keep working while it enforces budgets, guardrails, and smart routing. It includes Drop-in API, Budget enforcement, Smart Routing, and Guardrails, and is used with GitHub, Notion, Langfuse, Datadog, and Slack. Plans run Pay as you go $5 to start, Pro $50/mo or $500/yr in credits, and Enterprise custom.
Last verifiedHow we evaluate
At a glance
- NRouter is best for product and platform teams who need one governed gateway for many AI models.
- Pay as you go $5 to start; Pro $50/mo or $500/yr in credits; Enterprise Custom
- Yes — It offers an OpenAI-compatible endpoint and an API reference, with a drop-in API for accessing supported models through one key.
How the pricing actually compares
nRouter charges a flat 4% platform fee on pay-as-you-go credit purchases with no minimum fee (a $5 top-up costs $5.21), or 0% if you commit to the Pro plan at $50/mo or $500/yr, with the full amount landing as usable credits either way. For comparison, OpenRouter — the best-known gateway in this space — charges 5.5% on Stripe credit purchases with an $0.80 minimum, confirmed on OpenRouter's own FAQ page today. So nRouter's fee is genuinely lower at every tier we could verify, and free of the small-purchase penalty OpenRouter's minimum creates. New accounts get a one-time $10 bonus on a $5 first load ($15 to start).
Company behind it, and a name change worth knowing
nRouter is a d/b/a of CloudAct Inc., a California C-corp headquartered in Sunnyvale, founded in 2025 by Rama Surasani (Founder & CEO). The same company previously marketed this product as "NemoRouter" at nemorouter.ai — that domain now permanently redirects (308) to nrouter.ai/about, and NemoRouter's own marketing described the same May 2026 public-launch timeline as nRouter's current changelog. This reads as a rebrand of one product rather than two competing tools, not a red flag on its own, but worth knowing if you find NemoRouter referenced elsewhere — it's the same company and product under an earlier name.
Security and compliance posture
The vendor is unusually direct about where it stands: its security page states SOC 2 is "not yet" certified and that it "will not say 'certified' until the report is signed," with an audited report targeted for Q3 2026. It claims SOC 2-aligned controls in the meantime. Data is described as AES-256 encrypted at rest and TLS 1.2+ in transit, with 90-day request-log retention (longer on Enterprise), and data-residency options across the US, EU, UK, Canada, Australia, Singapore, and India. A HIPAA BAA is available on request. None of these compliance claims are independently audited yet — they're the vendor's own description of its controls, not a certification.
Enterprise features, without customer proof yet
The Enterprise tier lists SAML 2.0/OIDC SSO against Okta, Azure AD, or Google Workspace; an append-only audit trail of key, guardrail, budget, and team changes; role-based access control enforced via Postgres row-level security; and deployment options spanning managed multi-tenant (live today), dedicated single-tenant and VPC/BYO-cloud, with on-premise/air-gapped listed as roadmap only. A 99.9% uptime SLA is stated to apply on every tier, not just Enterprise. The enterprise page names no customers, case studies, or logos — there's nothing here to independently corroborate scale or reliability at this tier yet.
What's missing: independent evidence
We found no Trustpilot profile, no G2 or Capterra listing, and no Product Hunt reviews for nRouter. Its GitHub organization (github.com/nRouterAI) was created August 15, 2026 — five days before this review — and has zero public repositories and zero followers, so there's no code, issue tracker, or community activity to inspect there. The public changelog shows active shipping before launch (roughly every 3–5 days from March through early May 2026) but has had no entries since v1.0.1 on May 16, 2026, over three months before this review, despite the product having publicly launched only days earlier.
Model catalog: one internal inconsistency
The vendor's own homepage metadata describes Alibaba US (Qwen, DeepSeek) as "our largest catalog," but the live models page lists 47 total models broken down as 32 OpenAI, 7 Anthropic, 7 Qwen, and 1 other — meaning OpenAI, not Alibaba/Qwen, is actually the largest slice by a wide margin. This looks like stale marketing copy rather than a substantive claim, but it means the catalog-composition language on the homepage shouldn't be taken at face value.
How much does nRouter.ai cost?
| Plan | Price | What's included |
|---|---|---|
| Pay as you go | $5 to start |
|
| Pro | $50/mo or $500/yr in credits |
|
| Enterprise | Custom |
|
Frequently asked questions
What is nRouter.ai?
NRouter.ai is a managed LLM routing platform for product and platform teams that sits between your app and model providers, exposing an OpenAI-compatible endpoint so existing SDKs can keep working while it enforces budgets, guardrails, and smart routing. It includes Drop-in API, Budget enforcement, Smart Routing, and Guardrails, and is used with GitHub, Notion, Langfuse, Datadog, and Slack. Plans run Pay as you go $5 to start, Pro $50/mo or $500/yr in credits, and Enterprise custom.
How much does nRouter.ai cost? Is it free?
NRouter.ai has 3 paid plans: Pay as you go at $5 to start, Pro at $50/mo or $500/yr in credits, Enterprise at Custom.
What is nRouter.ai used for? Who is it for?
NRouter.ai is used for Drop-in API, Budget enforcement, and Smart Routing. It's built for Product teams shipping AI features that need one key, one bill, and per-key budgets, Agent and workflow builders, and Startups moving fast and want pay-as-you-go credits without feature gating.
Does nRouter.ai have an API and what does it integrate with?
It offers an OpenAI-compatible endpoint and an API reference, with a drop-in API for accessing supported models through one key. It integrates with Langfuse, Datadog, Slack, OpenAI SDK, AWS Bedrock, and 25 more.
Editor's read
Check whether the throughput and capacity terms on Pro are enough for your workload, since higher throughput and priority routing are tied to reserved capacity. If you need dedicated capacity, custom RPM/TPM/TPS, or SSO/SAML and SCIM, those are Enterprise items.
