Models Endpoint

Query real-time model catalog, context limits, input/output rates, and capabilities programmatically via GET /v1/models.

Toknix provides a programmatic catalog endpoint returning all 40 active models, their context windows, output bounds, per-million token rates, and feature flags.

GET /v1/models

bash
curl https://api.toknix.in/v1/models   -H "Authorization: Bearer $TOKNIX_API_KEY"

Response Schema

Each model object in the data array contains:

  • id: The exact string identifier to pass in requests.
  • object: Fixed to "model".
  • owned_by: Provider name ("Anthropic", "OpenAI", "Google", "xAI", "DeepSeek", "GLM", "Kimi").
  • context_window: Maximum total tokens (prompt + output).
  • max_output_tokens: Upper bound on completion tokens.
  • pricing: Object containing input_cost_per_mtok, output_cost_per_mtok, and cache_read_cost_per_mtok.
  • capabilities: Flags for streaming, tools, vision, and uncensored.

Example Response

json
{
  "object": "list",
  "data": [
    {
      "id": "glm-5.3-flash",
      "object": "model",
      "created": 1740000000,
      "owned_by": "GLM",
      "context_window": 1048576,
      "max_output_tokens": 67584,
      "pricing": {
        "currency": "USD",
        "input_per_mtok": 0.03,
        "output_per_mtok": 0.03,
        "cache_read_per_mtok": 0.003
      },
      "capabilities": {
        "streaming": true,
        "tools": true,
        "vision": false,
        "uncensored": false
      }
    },
    {
      "id": "kimi-k3",
      "object": "model",
      "created": 1740000000,
      "owned_by": "Kimi",
      "context_window": 1048576,
      "max_output_tokens": 67584,
      "pricing": {
        "currency": "USD",
        "input_per_mtok": 0.17,
        "output_per_mtok": 0.17,
        "cache_read_per_mtok": 0.04
      },
      "capabilities": {
        "streaming": true,
        "tools": true,
        "vision": false,
        "uncensored": false
      }
    }
  ]
}