Function

This API is used to obtain the AI engine and runtime of models.

URI

GET /v1/{project_id}/models/ai-engine-runtimes

**Table 1** Path Parameters
Parameter	Mandatory	Type	Description
project_id	Yes	String	Project ID. For details, see Obtaining a Project ID.

**Table 2** Query Parameters
Parameter	Mandatory	Type	Description
limit	No	String	Number of items displayed on each page.
offset	No	String	Offset, which is the position where the query starts. The value must be greater than or equal to 0.

Request Parameters

**Table 3** Request header parameters
Parameter	Mandatory	Type	Description
X-Auth-Token	Yes	String	User token

Response Parameters

Status code: 200

**Table 4** Response body parameters
Parameter	Type	Description
count	Integer	Total number of records that meet the search criteria when no pagination is performed
total_count	Integer	Number of records in the query result. If offset and limit are not configured, the values of count and total are the same.
engine_runtimes	Array of EngineAndRuntimesResponse objects	Engine runtime environment

**Table 5** EngineAndRuntimesResponse
Parameter	Type	Description
ai_engine	String	AI engine type. The options are as follows: TensorFlow PyTorch MindSpore XGBoost Scikit_Learn Spark_MLlib
runtimes	Array of strings	Runtime image, for example, pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64
request_mode	Array of strings	Request mode. The AI engine supports synchronous or asynchronous real-time services. sync (synchronous real-time service) async (asynchronous real-time service)
accelerators	Array of Accelerator objects	Accelerator cards that can be used by the AI engine
arch	Array of strings	AI engine architecture. The options are as follows: x86_64 aarch64
status_list	Array of strings	AI engine status. The options are as follows: normal: The AI engine is normal. deleted: The AI engine is deleted. deleting: The AI engine is being deleted.
image_source	Array of strings	Image source. The options are as follows: base_image uniform_image

**Table 6** Accelerator
Parameter	Type	Description
type	String	Accelerator card type. The options are as follows: npu gpu none
name	String	Accelerator card name, for example, a30
cuda_version	String	CUDA driver version
driver_version_section	String	Driver version set

Status code: 401

**Table 7** Response body parameters
Parameter	Type	Description
error_code	String	Error codes of ModelArts
error_msg	String	Error message

Status code: 403

**Table 8** Response body parameters
Parameter	Type	Description
error_code	String	Error codes of ModelArts
error_msg	String	Error message

Status code: 404

**Table 9** Response body parameters
Parameter	Type	Description
error_code	String	Error codes of ModelArts
error_msg	String	Error message

Example Requests

GET https://{endpoint}/v1/{project_id}/models/ai-engine-runtimes

Example Responses

Status code: 200

{
  "count" : 4,
  "total_count" : 4,
  "engine_runtimes" : [ {
    "ai_engine" : "TensorFlow",
    "runtimes" : [ "tf1.13-python3.6-cpu", "tf1.13-python3.6-gpu", "tf1.13-python3.7-cpu", "tf1.13-python3.7-gpu", "python3.6", "tf1.13-python3.7-aiflow-gpu", "tf1.13-python3.7-gpu-async", "tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64", "tensorflow_2.6.0-cuda_11.2-py_3.7-ubuntu_18.04-x86_64", "tensorflow_1.15.5-cuda_11.4-py_3.8-ubuntu_20.04-x86_64" ],
    "request_mode" : [ "sync", "sync", "sync", "sync", "sync", "sync", "async", "sync", "sync", "sync" ],
    "accelerators" : [ {
      "type" : "none"
    }, {
      "type" : "gpu",
      "cuda_version" : "cuda 10.2"
    }, {
      "type" : "none"
    }, {
      "type" : "gpu",
      "cuda_version" : "cuda 10.2"
    }, {
      "type" : "none"
    }, {
      "type" : "gpu",
      "cuda_version" : "cuda 10.2"
    }, {
      "type" : "gpu",
      "cuda_version" : "cuda 10.2"
    }, {
      "type" : "none"
    }, {
      "type" : "none"
    }, {
      "type" : "none"
    } ],
    "arch" : [ "x86_64", "x86_64", "x86_64", "x86_64", "x86_64", "x86_64", "x86_64", "x86_64", "x86_64", "x86_64" ],
    "status_list" : [ "normal", "normal", "normal", "normal", "normal", "normal", "normal", "normal", "normal", "normal" ],
    "image_source" : [ "base_image", "base_image", "base_image", "base_image", "base_image", "base_image", "base_image", "uniform_image", "uniform_image", "uniform_image" ]
  }, {
    "ai_engine" : "PyTorch",
    "runtimes" : [ "python3.6", "python3.7", "pytorch1.4-python3.7", "pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b", "pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64", "pytorch_1.8.2-cuda_11.1-py_3.7-ubuntu_18.04-x86_64" ],
    "request_mode" : [ "sync", "sync", "sync", "sync", "sync", "sync" ],
    "accelerators" : [ {
      "type" : "none"
    }, {
      "type" : "none"
    }, {
      "type" : "none"
    }, {
      "type" : "npu",
      "name" : "SNT9",
      "driver_version_section" : "C8x"
    }, {
      "type" : "none"
    }, {
      "type" : "none"
    } ],
    "arch" : [ "x86_64", "x86_64", "x86_64", "aarch64", "x86_64", "x86_64" ],
    "status_list" : [ "normal", "normal", "normal", "normal", "normal", "normal" ],
    "image_source" : [ "base_image", "base_image", "base_image", "uniform_image", "uniform_image", "uniform_image" ]
  }, {
    "ai_engine" : "MindSpore",
    "runtimes" : [ "mindspore_1.9.0-cann_6.0.1-py_3.7-euler_2.9.9-x86_64-snt3p-300i", "mindspore_1.7.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64", "mindspore_1.2.0-py_3.7-cuda_10.1-ubuntu_18.04-x86_64" ],
    "request_mode" : [ "sync", "sync", "sync" ],
    "accelerators" : [ {
      "type" : "none"
    }, {
      "type" : "none"
    }, {
      "type" : "none"
    } ],
    "arch" : [ "x86_64", "x86_64", "x86_64" ],
    "status_list" : [ "normal", "normal", "normal" ],
    "image_source" : [ "uniform_image", "uniform_image", "uniform_image" ]
  }, {
    "ai_engine" : "Custom",
    "runtimes" : [ ],
    "request_mode" : [ ],
    "accelerators" : [ ],
    "arch" : [ ],
    "status_list" : [ ],
    "image_source" : [ ]
  } ]
}

Status Codes

Status Code	Description
200	OK
401	Unauthorized
403	Forbidden
404	Not Found

Error Codes

See Error Codes.

Obtaining the Model Runtime