nRouter Abolishes Broker Markups with Flat List-Price Settlement and FinOps FOCUS 1.4 Tokenomics

SUNNYVALE, CA, August 19, 2026 — nRouter, the enterprise AI gateway built by CloudAct Inc., today announced a transformative overhaul of generative AI billing dynamics by establishing 0% per-token markup across all servable foundation models, coupled with an enterprise FinOps FOCUS 1.4 tokenomics reconciliation engine.

Across the software industry, third-party AI proxy brokers have routinely levied surcharges ranging from 5% to 15% on customer token consumption. As enterprise workloads scale into billions of monthly tokens, this markup creates an unsustainable tax on technological innovation and obscures true compute unit economics.

The Utility Model: Zero-Markup Settlement

nRouter treats artificial intelligence compute as a neutral public utility. Rather than penalizing customer usage volume through arbitrary token surcharges, nRouter bills every request at the upstream provider's published list price:

• Exact Upstream List Price: Customers pay the exact per-token rates published by Google Cloud Vertex AI, OpenAI, Anthropic, AWS Bedrock, and Azure.
• $0 Held on Blocked Threats: If a request is blocked by safety guardrails or compliance policies, the customer is charged exactly $0 and holds $0.
• Predictable Infrastructure Pricing: Customers pay a transparent platform subscription for gateway features, aligning nRouter's engineering incentives with driving down customer model spend through algorithmic routing efficiency.

FinOps FOCUS 1.4 Specification Compliance

To provide CFOs, procurement officers, and engineering managers with institutional-grade financial visibility, nRouter has adopted the FinOps Open Cost and Usage Specification (FOCUS 1.4).

The platform continuously performs automated 3-way reconciliation uniting:

1. Cloud Provider Invoices: Ingested billing line items from Google Cloud, AWS, and Microsoft Azure.
2. Upstream Direct Billing: Direct API usage statements from OpenAI and Anthropic.
3. Gateway Spend Telemetry: Immutable Postgres ledger logs capturing exact input, output, cached, and reasoning token counts per virtual key.

Virtual Keys & Granular Budget Guardrails

Engineering organizations can issue team-scoped virtual keys (sk-nrouter-...) bound to strict monthly or quarterly spend ceilings. When a team or autonomous agent reaches its allocated threshold, the gateway gracefully rejects further requests before financial overruns occur, eliminating the threat of rogue loops and billing surprises.

Executive Perspective

"Generative AI compute is rapidly becoming one of the largest line items in modern engineering budgets," stated Rama Surasani, Founder and CEO of nRouter. "Charging a percentage markup on model tokens is an outdated middleman business model. Enterprises demand mathematical pricing transparency, audit-ready FinOps integration, and the certainty that their gateway partner is actively helping them minimize compute costs, not inflate them."

Availability & Migration

The zero-markup pricing model and FOCUS 1.4 reconciliation tools are immediately active for all nRouter customers. Existing customers can configure their virtual key spend ceilings and download FOCUS-compliant CSV exports directly from the nRouter dashboard.

For media inquiries: sales@nrouter.ai. For FinOps enterprise consultations: sales@nrouter.ai.