RouterBase

About RouterBase

RouterBase provides a unified API that routes requests to over 200 AI models from providers including OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, and xAI, covering chat, image, video, and audio models. The API is OpenAI-compatible, allowing existing OpenAI SDK integrations to switch providers by changing the base URL, and official SDKs are also available for Python, Node.js, Go, and Ruby.

The service includes smart routing that selects a provider by price, latency, or availability per request, automatic fallback between providers on failure, unified billing across all connected providers, and usage analytics with spend tracking by model, key, or project that can be exported to CSV or queried via API. Pricing is pass-through token cost plus a flat platform fee, with real-time balance tracking and configurable per-key monthly spend caps. Bring-your-own-key support is available for OpenAI, Anthropic, and Google, letting users retain provider-side committed-spend discounts while paying RouterBase's flat platform fee, with traffic falling back to pooled provider access when a soft cap is reached.

Data handling options include zero-retention request forwarding where supported by the upstream provider, a toggle to disable dashboard logging, and processing regions in the EU, US, and APAC, with region pinning, per-key region locks, and signed DPAs available on Enterprise plans. Rate limits are configurable per key, starting at 60 requests per minute on the free tier and higher on Pro and Enterprise tiers, with webhooks that fire before a key hits its limit. The product targets development teams building applications on multiple LLM providers who want a single integration point instead of managing separate SDKs, accounts, and invoices for each provider.

Key features

• OpenAI-compatible API — drop-in replacement, just swap the base URL
• Access to 200+ AI models from OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek and xAI under one key
• Smart routing that auto-selects provider by price, latency or availability per request
• Automatic fallback between providers in milliseconds on failure
• Unified billing with one invoice across all providers
• Usage analytics with spend tracking by model, key or project, exportable to CSV or API

Frequently asked questions