Question 1

What is NemoRouter?

Accepted Answer

NemoRouter is a managed LLM gateway that gives teams one API key to access 114+ AI models from Anthropic (Claude), Google (Gemini), and OpenAI (GPT, o-series) live today — AWS Bedrock shipping next. It is OpenAI-compatible, includes built-in guardrails, real-time cost tracking, team management. Pricing is two simple plans plus Enterprise: Flat 4% pay-as-you-go with no minimum fee, or 0% on Pro ($50/mo, $500/yr).

Question 2

How is NemoRouter different from a public LLM gateway?

Accepted Answer

NemoRouter is purpose-built for teams, not individual developers. It combines the one-key-many-models experience of a public gateway with enterprise features that teams normally only get from self-hosted proxies: per-team budgets, org-wide RBAC, guardrails scoped per key, prompt management, and audit logs. All features are unlocked on every plan — plans only affect the platform fee (a flat 4% pay-as-you-go with no minimum fee, or 0% on Pro) and guaranteed RPM/TPM.

Question 3

What is the platform fee?

Accepted Answer

NemoRouter charges a platform fee on top of provider costs: a flat 4% on Credits (Pay as you go) with no minimum fee, 0% on Pro ($50/mo or $500/yr in credits), and 0% on Enterprise (custom). You always pay the full model price to the provider — the platform fee is separate and transparent.

Question 4

Which AI models does NemoRouter support?

Accepted Answer

NemoRouter currently serves 114+ models live across Anthropic (Claude Opus 4.8, Sonnet 4.6, Haiku 4.5), Google Vertex AI (Gemini 3.5 Flash, Gemini 3.1 Pro, Imagen, Veo, embeddings), and OpenAI (GPT-5.5, GPT-4o, o-series, text-embedding-3). AWS Bedrock is shipping next. New models are added within days of general availability.

Question 5

Is NemoRouter OpenAI-compatible?

Accepted Answer

Yes. NemoRouter exposes an OpenAI-compatible REST API at https://api.nemorouter.ai/v1. Existing code using the OpenAI SDK works by changing two lines: the base URL and the API key. All features (routing, guardrails, observability, budgets) work through the standard chat completions endpoint — no custom SDK required.

Question 6

How does cost tracking work?

Accepted Answer

NemoRouter reports the exact provider cost on every request (x-nemo-request-cost) and charges your credit balance. You get per-request, per-model, per-team, and per-key cost breakdowns in real time via the dashboard and the /usage endpoint. Budgets can be enforced at any of those scopes and trigger 429s when exceeded.

Question 7

What guardrails does NemoRouter provide?

Accepted Answer

NemoRouter runs pre-call, during-call, and post-call guardrails from providers like Aporia, Guardrails AI, Bedrock Guardrails, Azure Content Safety, and Lasso. Guardrails are configured per key, per team, or per org. All providers are unlocked on every plan — you do not need to upgrade to enable safety controls.

Question 8

Is there a free tier or free credits?

Accepted Answer

Every new account gets $10 in free credits to start with no credit card required, so you can test any model immediately. There is no recurring free tier for usage — you pay only for the credits you use — but every product feature (guardrails, prompts, A/B tests, team RBAC, SSO, audit logs, observability) is unlocked on every plan, including pay-as-you-go.

Question 9

How do I migrate from OpenAI or OpenRouter to NemoRouter?

Accepted Answer

Migration takes two lines. Because NemoRouter is OpenAI-compatible, you change the base URL to https://api.nemorouter.ai/v1 and swap your API key for a NemoRouter key. No SDK change, no request reformatting — existing OpenAI, LangChain, LlamaIndex, and Vercel AI SDK code keeps working. Model names map directly to the same provider models.

Question 10

Does NemoRouter support streaming and function calling?

Accepted Answer

Yes. Streaming uses Server-Sent Events with the standard OpenAI streaming protocol (set stream: true). Function calling / tool use is supported on every model that exposes it upstream, using the same tools and tool_choice parameters as the OpenAI API. Embeddings and image generation are also available through the standard endpoints.

Feature	NemoRouter	OpenRouter	Kong AI Gateway	Portkey	Helicone
Pricing Model	Credits — 0–4% platform fee	5.5% markup	Enterprise license	Per-feature SaaS	Per-event metering
Managed Infrastructure	Included	Included	Self-host or Konnect SaaS	Included	Included
All Features Unlocked	Every tier	Limited	Enterprise tier only	Paid tier only	Paid tier only
Team & Org Management	Included	—	Workspaces (paid)	Paid tier only	Paid tier only
Guardrails & Safety	Built-in	—	AI plugins (basic)	Paid tier only	—
Budget Controls	Per key / team / org	Basic	Rate limit only	Paid tier only	Basic
Audit Logging	Included	—	Enterprise only	Paid tier only	—
OpenAI-Compatible API	Yes	Yes	Via plugin	Yes	Yes
Setup Time	Sign up & call	Sign up & call	Provision gateway & plugins	Sign up & call	Sign up & call

Every model. One API key.
Total control.

Every request runs through one gateway.

Every model, one API.

Everything you need. Nothing you don't

Smart Routing

Guardrails

One API Key

Spend Tracking

Budget Controls

Prompt Templates

Team Management

Observability

Model Catalog

One bill. No feature gating

Credits

Pro

Enterprise

Why teams choose NemoRouter

One API key.
Every model. Three tiers.

Every model. One API key.Total control.You can't overspendby accident.Guardrails on by default.Not a tier you buy.Ten providers. One bill.Zero surprises.See cost against volume,in real time.