get_model_schema

Generation

Get Model Schema

Get the input parameter schema for any Scenario model. Call this before run_model to discover accepted fields, types, defaults, and constraints. Args: - model_id: required — model to inspect - team_id, project_id: required for OAuth callers - response_format: 'json' (default) or 'markdown' Returns: model metadata plus a parameters array with field names, types, descriptions, required flags, defaults, and allowed values when available. Examples: - "What parameters does this model accept?" -> model_id="model_google-gemini-3-1-flash" - "Inspect this model before generating" -> call get_model_schema, then build run_model parameters from the response Don't use when: You need to discover models by task or name. Prefer search or recommend instead.

read-onlyidempotentopen-world

Parameters

Name	Type	Required	Description
model_id	string	✓	Model ID to get the schema for (e.g. 'model_google-gemini-3-1-flash').
team_id	string	—	Team ID. Required if user belongs to multiple teams.
project_id	string	—	Project ID to scope the operation to.
response_format	enum(json \| markdown)	json	Output format: 'json' for structured data, 'markdown' for human-readable text.

Example Request

JSON

{
  "model_id": "model_google-gemini-3-1-flash",
  "team_id": "team_abc123",
  "project_id": "proj_xyz789"
}

Example Response

JSON

{
  "model_id": "model_staging_google-gemini-3-1-flash",
  "name": "Gemini 3.1 Flash 🍌",
  "capabilities": [
    "txt2img",
    "img2img"
  ],
  "parameters": [
    {
      "name": "prompt",
      "type": "string",
      "description": "Text prompt for image generation/editing",
      "required": true,
      "max_length": 4096
    },
    {
      "name": "referenceImages",
      "type": "file_array",
      "description": "Reference images for style or content guidance",
      "required": false,
      "max_length": 14,
      "cost_impact": true
    },
    {
      "name": "aspectRatio",
      "type": "string",
      "description": "Aspect ratio for the generated image",
      "required": false,
      "default": "auto",
      "allowed_values": [
        "21:9",
        "16:9",
        "3:2",
        "4:3",
        "5:4",
        "1:1",
        "4:5",
        "3:4",
        "2:3",
        "9:16",
        "auto"
      ]
    },
    {
      "name": "resolution",
      "type": "string",
      "description": "Resolution for the generated image",
      "required": false,
      "default": "1K",
      "allowed_values": [
        "512",
        "1K",
        "2K",
        "4K"
      ],
      "cost_impact": true
    },
    {
      "name": "useGoogleSearch",
      "type": "boolean",
      "description": "Use Google Search to find more information about the prompt",
      "required": false,
      "default": false,
      "cost_impact": true
    },
    {
      "name": "numOutputs",
      "type": "number",
      "description": "Number of images to generate",
      "required": false,
      "default": 1,
      "min": 1,
      "max": 4,
      "cost_impact": true
    },
    {
      "name": "seed",
      "type": "number",
      "description": "Use a seed for reproducible results. Leave blank to use a random seed.",
      "required": false,
      "min": 0,
      "max": 2147483647
    }
  ]
}

Common Use Cases

Discover all accepted parameters before calling run_model
Check allowed aspect ratios and resolutions for a specific model
Find required fields to avoid validation errors at generation time
Understand cost-impacting parameters before running an expensive job