Docs
Compatibility
Which OpenAI-compatible endpoints the gateway serves, SDK versions tested, error shapes, and rate-limit behavior.
Base URL
All requests go to the same origin: https://easyairoute.com/v1. Replace https://api.openai.com/v1 with this URL in any OpenAI SDK or tool — nothing else changes.
Endpoints
The gateway serves the core OpenAI endpoints. Model availability depends on which channels the operator has enabled; the models list endpoint reflects exactly what your key can call.
POST /v1/chat/completions # chat, tools, vision, JSON mode
POST /v1/images/generations # text-to-image (billed per item)
POST /v1/images/edits # image-to-image editing
GET /v1/models # list models available to your keySDKs and tools
Anything that lets you set a custom OpenAI base URL works: the official Python and Node SDKs, LangChain, LlamaIndex, curl, and most GUI clients. Tested against openai-python >= 1.x and openai-node >= 4.x.
Errors
Error bodies follow the OpenAI format: { "error": { "message", "type", "code" } }. Common cases: 401 means an invalid or deleted key; 403 means the model is not enabled for your group; 429 means quota exhausted or rate limited. For billing corrections, keep the request ID and contact support.
Rate limits
Per-key limits are configurable by the operator; there is no fixed global cap. If you need higher throughput for production workloads, contact the operator to raise your key's limit.