One OpenAI- and Anthropic-compatible API for text, image and video models from leading AI labs. One key, one prepaid USD balance, published per-token prices.
Switching between AI models shouldn't mean rewriting your integration. Most teams now run several, but each provider still means another SDK, another key and another invoice.
BoostRail gives developers one API for text, image and video models from Anthropic, OpenAI, Google, DeepSeek and other labs. It works with the OpenAI and Anthropic SDKs unchanged: point the base URL at BoostRail, use one key, and pick a model by its id. If a route fails before the response starts, the request moves to another route of the same model, never to a different model.
Our stack
Works with the OpenAI and Anthropic SDKs and any HTTP client, plus Claude Code, Codex CLI, Cline, LiteLLM, LibreChat and the Vercel AI SDK. Streaming over server-sent events. Usage is metered per token in USD against one prepaid balance, with prices published per model.
