Instagram

Docs

Models

List the models available to your API key, or retrieve one by ID.

Two OpenAI-compatible endpoints for discovering what you can call. Both authenticate with your elyxir_ API key.

List models

GET https://api.elyxir.ai/v1/models

Returns every model your key can use, in the standard OpenAI list shape. What you get back depends on your plan — higher plans see more of the catalog.

curl https://api.elyxir.ai/v1/models \
  -H "Authorization: Bearer elyxir_your_api_key"

Response

{
  "object": "list",
  "data": [
    {
      "id": "elyxir-alvin",
      "object": "model",
      "created": 1234567890,
      "owned_by": "elyxir"
    },
    {
      "id": "gpt-4o",
      "object": "model",
      "created": 1234567890,
      "owned_by": "openai"
    }
  ]
}

elyxir-alvin appears in this list as a virtual model. It has no provider of its own — requesting it hands the call to Alvin's router, which picks a real model per request.

Retrieve a model

GET https://api.elyxir.ai/v1/models/{model}

curl https://api.elyxir.ai/v1/models/gpt-4o \
  -H "Authorization: Bearer elyxir_your_api_key"

{
  "id": "gpt-4o",
  "object": "model",
  "created": 1234567890,
  "owned_by": "openai"
}

Errors

Status	Meaning
401	Invalid or revoked API key
404	Model not found (retrieve only)

Notes

Call /v1/models at startup rather than hardcoding IDs. The catalog changes as providers ship and retire models, and a model that worked last month may not resolve today.

Chat Completions — the main inference endpoint
Supported models — full catalog with pricing