Models

GET /v1/models

Auth: none required. Returns the catalog of model names the gateway accepts. The codex/… tiers and flagship/mini aliases resolve to your account’s current Codex models at request time, so this list stays stable while OpenAI rotates the underlying names.

curl https://api.proxyllm.ai/v1/models

Response

OpenAI list shape:

{
  "object": "list",
  "data": [
    {
      "id": "gpt-5",
      "object": "model",
      "created": 1751500000,
      "owned_by": "proxyllm"
    },
    {
      "id": "gpt-4o-mini",
      "object": "model",
      "created": 1751500000,
      "owned_by": "proxyllm"
    },
    {
      "id": "claude-sonnet-4-6",
      "object": "model",
      "created": 1751500000,
      "owned_by": "proxyllm"
    },
    {
      "id": "codex/flagship",
      "object": "model",
      "created": 1751500000,
      "owned_by": "codex"
    },
    {
      "id": "codex/mini",
      "object": "model",
      "created": 1751500000,
      "owned_by": "codex"
    },
    {
      "id": "bridge/claude-code",
      "object": "model",
      "created": 1751500000,
      "owned_by": "bridge"
    }
  ]
}

Which entries actually serve depends on the routing key’s lanes; use GET /v1/key to see what a specific key can reach.

Errors

None specific to this endpoint.