Home / Technology / Ramp Launches AI Model Router Service
Ramp Launches AI Model Router Service
20 Aug
Summary
- Ramp introduced Router, an API service for switching AI language models.
- The service offers access to models from multiple providers like OpenAI and Anthropic.
- Router provides routing strategies based on user preferences and performance benchmarks.
Corporate expense management platform Ramp has entered the AI inference market with the launch of its new service, Router. Available through an API, Router enables users and companies to access and seamlessly switch between multiple large language models. The service currently offers models from prominent providers such as OpenAI, Anthropic, and Nvidia.
Router features several routing strategies, allowing customers to direct AI queries based on their specific needs. These strategies include prioritizing certain usage tiers or letting the service select the best model using user-defined benchmarks. A dashboard provides insights into token spend, cost, and latency.
The service is currently only available in the United States and is free to use through the end of 2026, though users will still incur AI model inference costs. Ramp has also provided a $26 credit for new users upon launch. The company has an opt-out data retention policy, recording model inputs and outputs for one year by default, with measures to remove personally identifiable information.
