Cut.ProDocs
Generate with AI: models

List the models worth starting with

The same list as GET /models, cut to a handful of strong models per kind: what to offer when the user names no model.

Authorization

apiKey
X-Api-Key<token>

In: header

Query Parameters

kind?string

Value in

  • "image"
  • "video"
  • "speech"
  • "music"
  • "audio"
  • "avatar"
  • "model3d"
  • "tool"
  • "videoedit"
  • "text"
  • "embedding"
endpoint?string

Value in

  • "generations"
  • "chat_completions"
  • "embeddings"
q?string

Part of the id or the name.

Lengthlength <= 64
id?string

One model's id.

Lengthlength <= 128

Response Body

application/json

GET
/models/featured
curl -X GET "https://example.com/models/featured"
{  "models": [    {      "id": "string",      "name": "string",      "maker": "string",      "kind": "image",      "endpoint": "generations",      "featured": true,      "pricing": [        {          "unit": "image",          "credits": 0,          "mode": "text",          "resolution": "string",          "silent_credits": 0        }      ],      "priced_by_request": true,      "parameters": {        "properties": {          "property1": {            "type": "string",            "enum": [              "string"            ],            "default": "string",            "minimum": 0,            "maximum": 0,            "maxItems": 0,            "items": "string",            "description": "string"          },          "property2": {            "type": "string",            "enum": [              "string"            ],            "default": "string",            "minimum": 0,            "maximum": 0,            "maxItems": 0,            "items": "string",            "description": "string"          }        },        "required": [          "string"        ]      },      "supported_parameters": [        "string"      ],      "context_length": 0,      "max_output_tokens": 0,      "input_modalities": [        "string"      ],      "modes": [        "text"      ],      "resolutions": [        "string"      ],      "seconds": [        0      ],      "mode_seconds": [        {          "mode": "text",          "seconds": [            0          ]        }      ],      "aspect_ratios": [        "21:9"      ],      "audio": "always",      "max_images": 0,      "max_audios": 0,      "max_video_seconds": 0,      "refuses": [        "real_faces"      ],      "voices": [        {          "id": "string",          "name": "string",          "style": "string",          "gender": "female",          "age": "young",          "language": "string",          "preview_url": "string",          "portrait_url": "string"        }      ],      "quality": 0,      "typical_seconds": 0,      "success_rate": 0    }  ],  "presets": {    "video": [      {        "id": "fast",        "model": "string",        "resolution": "string"      }    ],    "image": [      {        "id": "fast",        "model": "string",        "resolution": "string"      }    ],    "speech": [      {        "id": "fast",        "model": "string",        "resolution": "string"      }    ]  },  "prompt_max": 0,  "max_running": 0,  "improve_credits": 0}

List every AI model GET

Every model CutPro runs, of every kind: chat models (Claude, GPT, Gemini, DeepSeek, Qwen, Llama and more), embeddings, images, videos, voices, music, sound effects, talking avatars, 3D models and image and video tools, with what each takes, where it runs and its price in credits. Prices can change at any time: read them from here, or get the exact total of a generation from `POST /generations/quote`. Narrow it with `kind`, `endpoint`, `q` or `id`; `GET /models/featured` lists the ones worth starting with.

Upload a reference image POST

Sends one image as `multipart/form-data`, to animate it, use it as a first or last frame, or borrow a person, product or style from it. It goes to the workspace library and counts toward its storage. Pass the returned `id` in `image_ids` of `POST /generations`. An image CutPro generated needs no upload: its `output.image_id` works the same.