{
  "openapi": "3.0.0",
  "info": {
    "title": "WAVE Inference",
    "version": "1.0.0",
    "description": "One OpenAI-compatible endpoint fronting thirteen providers with measured routing, automatic failover, and per-token metering."
  },
  "servers": [
    {
      "url": "https://inference.wave.online"
    }
  ],
  "paths": {
    "/v1/chat/completions": {
      "post": {
        "summary": "Create a chat completion",
        "description": "OpenAI-compatible chat completion. Auth: Authorization: Bearer <key> (virtual keys carry budgets + model allowlists).",
        "responses": {
          "200": {
            "description": "OpenAI-shaped completion response"
          }
        }
      }
    },
    "/v1/models": {
      "get": {
        "summary": "List available models",
        "description": "Every model carries its measured health and price.",
        "responses": {
          "200": {
            "description": "List of models"
          }
        }
      }
    },
    "/v1/embeddings": {
      "post": {
        "summary": "Create embeddings",
        "description": "OpenAI-compatible embeddings endpoint.",
        "responses": {
          "200": {
            "description": "Embedding vectors"
          }
        }
      }
    },
    "/health/liveliness": {
      "get": {
        "summary": "Liveliness probe",
        "description": "Public status probe.",
        "responses": {
          "200": {
            "description": "\"I'm alive!\""
          }
        }
      }
    }
  }
}