An OpenAI-compatible API matches the request and response shape of the OpenAI API. That means you keep your existing SDK and code and only change the base URL and key. Infere exposes every provider through this shape, so switching models does not mean rewriting integrations.
Why it matters
Compatibility removes migration cost. If the API matches OpenAI's shape, existing SDKs, tools, and code work with only a base URL and key change.
How it works
The gateway accepts the standard chat completions request and returns the standard response, translating to and from each provider's native format internally. Your application never sees provider-specific payloads.