Models API

Browse the catalog of models available for deployment. All routes are mounted under /api/v1/deployable-models and require an authenticated session.

List Deployable Models

GET /api/v1/deployable-models

Returns the catalog of models available for deployment.

Query parameters:

Parameter	Type	Default	Description
`enabled_only`	boolean	true	Return only enabled models
`model_type`	string	LLM	Filter by model type
`provider`	string	—	Filter by provider
`q`	string	—	Free-text search

Response:

{
  "items": [
    {
      "id": "dm_abc",
      "model_name": "qwen2.5-instruct",
      "display_name": "Qwen2.5 Instruct",
      "provider": "Alibaba",
      "family": "Qwen2.5",
      "model_type": "LLM",
      "description": "Instruction-tuned Qwen2.5 for chat and text generation.",
      "tags": ["chat", "multilingual", "code"],
      "enabled": true,
      "sort_order": 10,
      "created_at": "2025-01-01T00:00:00Z",
      "updated_at": "2025-06-01T00:00:00Z"
    }
  ]
}

Get Deployable Model

GET /api/v1/deployable-models/{model_id}

Query parameters:

Parameter	Type	Default	Description
`enabled_only`	boolean	true	Only return the model if it is enabled

Returns a single deployable model (same schema as the items above). Responds 404 Not Found if the model does not exist or is not available to the current user.