Payments biz Stripe has finalized an agreement to acquire AI model routing platform OpenRouter for at least $7 billion. This move positions Stripe as a mediator of AI token sales, capturing the foundational plumbing of LLM infrastructure as metered pricing becomes a focus for the AI industry.
Patrick Collison recently said that metered pricing is the native business model of the AI era. By acquiring OpenRouter, Stripe now sees where the tokens are going. The transaction follows a rise for OpenRouter, which in May closed a $113 million funding round at a post-money valuation of $1.3 billion.
From Financial Ledgers to Inference Routing
Founded in 2023 by Alex Atallah and Louis Vichy, OpenRouter established itself as an integration layer for developers looking to avoid lock-in. Instead of writing custom API code for every discrete frontier model, developers point their applications at a single OpenAI Chat Completions-compatible endpoint. The gateway handles routing, failover, and unified billing across every model it lists.
That architectural simplicity fueled growth. By December 2025, OpenRouter reported serving more than 5 million developers routing traffic to more than 300 models across more than 70 providers. Today, those numbers have expanded to more than 10 million developers and 80 providers. SiliconANGLE and Sacra noted that as of March 2026, OpenRouter’s annualized revenue sat around $50 million, while The Information shared that it climbed to approximately $140 million by mid-2026, resulting in a trailing revenue multiple of roughly 50x against the $7 billion buyout valuation.
The strategic puzzle pieces began falling into place for Stripe in late 2025 with its acquisition of Metronome, a real-time usage tracking and billing platform built specifically for AI companies. Metronome gave Stripe visibility into how much AI consumption was being billed. OpenRouter currently acts as the overarching routing mechanism positioned above all market models, determining the destination for each query sent by developers.
The Shift Toward Open-Weight Models
The economics of LLM consumption are undergoing a shift. While proprietary frontier labs like OpenAI and Anthropic have focused on building their own brands and steering customers toward their tooling, open-weight models are gaining share. Data published by Vercel’s AI Gateway indicates that open-weight model token spend grew from 11 percent in April to 36 percent by July.
Chinese AI laboratories, including releases such as GLM 5.2 and Kimi K3, have captured a meaningful share of workloads historically owned by closed-weight labs. This presents an operational hurdle for Stripe.
- OpenRouter Scale: More than 10 million developers and 80 providers.
- Traffic Volume: Serving more than 1 trillion tokens per day in October 2025.
- Valuation Trajectory: Rose from a $1.3 billion post-money valuation in May to a reported agreement for at least $7 billion.
Neutrality and Structural Contradictions
Industry observers have raised questions about whether OpenRouter can preserve its neutrality under Stripe’s roof. Akhil Verghese, founder and CEO of Krazimo, noted that OpenRouter is to AI models what Stripe has been to financial infrastructure, emphasizing that maintaining impartial routing is critical to developer trust. If the platform begins favoring specific providers, the resulting loss of trust could be harmful.
Furthermore, managing a gateway where Chinese-origin models command the majority of developer tokens introduces operational wrinkles, as these are subject to China’s National Intelligence Law. Stripe has not said how it intends to resolve these structural contradictions.
When proprietary labs guide clients toward API-based billing ahead of public offerings, they run the danger of pushing customer traffic toward cheaper open-weight alternatives. Stripe is no longer just processing the credit card transactions that pay for software. Through OpenRouter, the company now sits at the routing layer of global intelligence.
