Ramp launches its own AI model router called Router

Ramp, an enterprise expense management platform, is hard at work following Stripe in setting up a toll booth for AI inference.

Ramp on Wednesday night launched its own AI model routing service called Router, allowing users and businesses to use and switch between different large-scale language models through an API. The company says it has been using routers it built in-house for its AI-enabled needs for the past three years.

This service is currently available only in the United States. Free to use for the remainder of 2026 (users will still have to pay AI model inference costs). It also comes with a $26 credit launch offer. The company did not say how much the service will cost next year.

Router works similarly to OpenRouter, but the latter offers more AI model options than Ramp’s current offering.

Router provides access to models from OpenAI, Anthropic, DeepSeek, Moonshot, Minimax, Nvidia, xAI, and Z.ai. It also provides several “strategies” that allow customers to route AI requests to models based on their preferences. For example, one allows the user to set preferences for a model provider’s flex usage tier, and the other allows the router to select a model to route queries to based on up to three user-specified benchmarks. Users can also route only difficult problems to more expensive models or easily test them without switching models.

Users also have a dashboard where they can view token usage, costs, latency, fallback attempts, and other details.

Notably, Router has an opt-out data retention policy, and model inputs, outputs, and tool calls are logged for one year by default, but the company says it removes “personally identifiable information before using that content to improve its products.”

For Ramp, entering the model routing business offers two opportunities. One is to tap into the fast-growing AI inference market, and the other is to provide existing customers with a model routing service that fits well with their existing products, including AI token usage monitoring and token spend management.

And if Router proves as attractive as OpenRouter for the model testing space, Ramp could also build long-term relationships with AI labs and inference providers around the world. This could help Ramp, which raised $750 million in June at a $44 billion valuation, gain new customers and new outlets to sell its expense management products.