Skip to main content
GET
/
batches
/
models
List batch model capabilities
const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.phaseo.app/v1/batches/models', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
{
  "object": "list",
  "data": [
    {
      "model": "<string>",
      "name": "<string>",
      "status": "<string>",
      "input_types": [
        "<string>"
      ],
      "output_types": [
        "<string>"
      ],
      "supported_params": [
        "<string>"
      ],
      "supported_parameters": [
        "<string>"
      ],
      "supported_params_detail": {},
      "supported_parameters_detail": {},
      "providers": [
        {
          "id": "<string>",
          "supported_params": [
            "<string>"
          ],
          "supported_parameters": [
            "<string>"
          ],
          "supported_params_detail": {},
          "supported_parameters_detail": {}
        }
      ],
      "pricing": {}
    }
  ]
}
Returns active models that can be used with the Batch API. Use this endpoint before creating a batch when you need to discover provider-specific batch constraints such as:
  • allowed endpoint values for batch input rows
  • supported completion_window values
  • model/provider-specific parameters exposed through supported_params_detail and supported_parameters_detail
The response is a standard list envelope. Each item includes:
  • model
  • name
  • status
  • input_types and output_types
  • supported_params
  • supported_parameters
  • supported_params_detail
  • supported_parameters_detail
  • providers, each with provider-level supported parameter detail
  • pricing
supported_parameters is the OpenRouter-compatible alias for supported_params. supported_params_detail and supported_parameters_detail are aliases for the same structured metadata. New clients should prefer supported_parameters and supported_parameters_detail; existing clients can continue reading supported_params and supported_params_detail. You can filter models by supported batch parameter with the params query parameter. Repeat it or pass a comma-separated list. The filter accepts canonical names and aliases from supported_parameters_detail, so params=request_endpoint can match models whose canonical parameter is endpoint, and params=window can match completion_window. 
curl "https://api.phaseo.app/v1/batches/models?params=request_endpoint,window" \
  -H "Authorization: Bearer $AI_STATS_API_KEY"

{
  "object": "list",
  "data": [
    {
      "model": "openai/gpt-5-mini",
      "status": "active",
      "supported_params": ["endpoint", "completion_window"],
      "supported_parameters": ["endpoint", "completion_window"],
      "supported_parameters_detail": {
        "endpoint": {
          "supported": true,
          "values": ["/v1/responses", "/v1/chat/completions"],
          "providers": ["openai"]
        },
        "completion_window": {
          "supported": true,
          "values": ["24h"],
          "providers": ["openai"]
        }
      }
    },
    {
      "model": "openai/text-embedding-3-small",
      "status": "active",
      "supported_params": ["endpoint", "completion_window"],
      "supported_parameters": ["endpoint", "completion_window"],
      "supported_parameters_detail": {
        "endpoint": {
          "supported": true,
          "values": ["/v1/embeddings"],
          "providers": ["openai"]
        },
        "completion_window": {
          "supported": true,
          "values": ["24h"],
          "providers": ["openai"]
        }
      }
    },
    {
      "model": "openai/gpt-image-1.5",
      "status": "active",
      "supported_params": ["endpoint", "completion_window"],
      "supported_parameters": ["endpoint", "completion_window"],
      "supported_parameters_detail": {
        "endpoint": {
          "supported": true,
          "values": ["/v1/images/generations", "/v1/images/edits"],
          "providers": ["openai"]
        },
        "completion_window": {
          "supported": true,
          "values": ["24h"],
          "providers": ["openai"]
        }
      }
    },
    {
      "model": "openai/omni-moderation",
      "status": "active",
      "supported_params": ["endpoint", "completion_window"],
      "supported_parameters": ["endpoint", "completion_window"],
      "supported_parameters_detail": {
        "endpoint": {
          "supported": true,
          "values": ["/v1/moderations"],
          "providers": ["openai"]
        },
        "completion_window": {
          "supported": true,
          "values": ["24h"],
          "providers": ["openai"]
        }
      }
    },
    {
      "model": "openai/sora-2",
      "status": "active",
      "supported_params": ["endpoint", "completion_window"],
      "supported_parameters": ["endpoint", "completion_window"],
      "supported_parameters_detail": {
        "endpoint": {
          "supported": true,
          "values": ["/v1/videos"],
          "providers": ["openai"]
        },
        "completion_window": {
          "supported": true,
          "values": ["24h"],
          "providers": ["openai"]
        }
      }
    }
  ]
}

Authorizations

​
Authorization
string
header
required

Bearer token authentication

Query Parameters

​
params
string[]

Optional supported-parameter filter. Repeat the parameter or pass a comma-separated list. Canonical names and aliases are accepted, for example endpoint, request_endpoint, completion_window, or window.

Response

Batch model capability list

​
object
string
Example:

"list"

​
data
object[]
Last modified on June 11, 2026