Help Center/ ModelArts/ API Reference/ Model Training/ Training Job Management (New Version)/ Obtaining Fine-tuning Training Job Details

Updated on 2026-09-01 GMT+08:00

View PDF

Obtaining Fine-tuning Training Job Details

Function

This API is used to obtain details about a specified fine-tuning training job on ModelArts.

It is used when you need to view the details about a specified training job, such as the running status and configuration information. Before using this API, ensure that you have obtained the training job ID and have the permission to view job details. After the query is complete, the platform returns the details about the training job, including the job status, configuration, and logs. If the training job ID does not exist or you do not have the operation permission, the API will return an error message.

Debugging

You can debug this API through automatic authentication in API Explorer or use the SDK sample code generated by API Explorer.

Authorization Information

Each account has all the permissions required to call all APIs, but IAM users must be assigned the required permissions.

If you are using role/policy-based authorization, see Permissions Policies and Supported Actions for details on the required permissions.
If you are using identity policy-based authorization, no identity policy-based permission required for calling this API.

URI

GET /v2/{project_id}/training-jobs/{training_job_id}/ft-detail

**Table 1** Path Parameters
Parameter	Mandatory	Type	Description
project_id	Yes	String	Definition: Project ID. For details, see Obtaining a Project ID and Name. Constraints: N/A Range: The value can contain 1 to 64 characters. Only letters, digits, and hyphens (-) are allowed. Default Value: N/A
training_job_id	Yes	String	Definition: Training job ID. For details, see Obtaining Training Jobs. Constraints: N/A Range: N/A Default Value: N/A

Request Parameters

None

Response Parameters

Status code: 200

**Table 2** Response body parameters
Parameter	Type	Description
project_id	String	Project ID.
task_id	String	Training task ID.
task_name	String	Training task name.
task_desc	String	Training job description.
metadata	JobMetadataResponse object	Definition: Training job metadata.
spec	SpecResponse object	Definition: Training job specifications.
model_asset_id	String	Model ID.
model_type	String	Definition Model type. The value can be TextGeneration (text generation) or ImageUnderstanding (image understanding). Range TextGeneration or ImageUnderstanding
model_source	String	Model source
train_type	String	Definition Training type. The value can be SFT (full fine-tuning), PRETRAIN (pre-training), LORA (LoRA fine-tuning), DPO (DPO reinforcement learning), or RFT (RFT reinforcement learning). Range SFT (full fine-tuning), PRETRAIN (pre-training), LORA (LoRA fine-tuning), DPO (DPO reinforcement learning), or RFT (RFT reinforcement learning) SFT
checkpoint_config	String	Resumable training configuration.
task_parameters	String	Training task parameters.
create_time	Long	Creation time.
update_time	Long	Time when a training task is updated. The value is updated when the task status is updated or the description is modified.
train_process	Double	Training task progress.
datasets_config	Array of DatasetConfig objects	Dataset configuration of the training task.
status	Status object	Definition: Training job status.
auto_publish_config	String	Automatic publish configuration
asset_code	String	Model asset name
asset_name	String	Asset name
asset_desc	String	Model asset description
asset_series	String	Model series
asset_version	String	Asset version
asset_type	String	Asset type
asset_source	String	Asset source
asset_group_id	String	Asset group ID
sub_asset_type	String	Asset subtype
category	String	Asset category
api_version	String	Asset API version
root_asset_id	String	Root asset ID.
train_cost_time	Long	Training task duration
workspace_id	String	Name of the workspace to which the task belongs
user_id	String	User ID
user_name	String	Username
pool_type	String	Resource pool type
pool_id	String	Resource pool ID
pool_node_count	String	Number of used resource pool instances
flavor_id	String	Used resource pool PUs
priority	Integer	Priority
training_info	String	Estimated training duration
train_output_path	String	Definition: output path of training artifacts, for example, obs://yyy/test/. Range: N/A
asset_capabilities	Array of strings	Training model type
continue_task	ContinueTask object	Resumable training task information

**Table 3** JobMetadataResponse
Parameter	Type	Description
id	String	Definition: Training job ID, which is generated and returned by ModelArts after a training job is created. Range: N/A
name	String	Definition: Name of a training job. Range: The value must contain 1 to 64 characters consisting of only digits, letters, underscores (_), and hyphens (-).
workspace_id	String	Definition: Workspace where a specified job is located. Range: N/A
description	String	Definition: Definition of a training job. Range: N/A
create_time	Long	Definition: Time when a training job was created, in milliseconds. The value is generated and returned by ModelArts after a training job is created. Range: N/A
user_name	String	Definition: Username for creating a training job. The username is generated and returned by ModelArts after a training job is created. Range: N/A
annotations	Map<String,String>	Definition: advanced function configuration of a training job. The key is the function switch/configuration name, and the value is a string. (Even if the value is a number or Boolean value, it is represented by a string, for example, "true" or "3".)
training_experiment_reference	TrainingExperimentResp object	Definition: Training experiment parameters.

**Table 4** TrainingExperimentResp
Parameter	Type	Description
name	String	Definition: Experiment name. Range: The value can contain a maximum of 128 characters. Special characters are not allowed.
id	String	Parameter Explanation: Experiment ID. Value Range: N/A.
serial_number	String	Definition: sequence number of the current training job in the training experiment to which the job belongs. The default value is 0.

**Table 5** SpecResponse
Parameter	Type	Description
resource	Resource object	Definition: Resource flavor of a training job. Select either flavor_id or pool_id and flavor_id.
volumes	Array of JobVolumeResp objects	Definition: Mounting volume information of a training job.
log_export_path	LogExportPathResp object	Definition: Log output of a training job.
schedule_policy	SchedulePolicyResp object	Definition: Scheduling policy of a training job.
custom_metrics	Array of CustomMetrics objects	Definition: Metric collection configuration.
output_model	OutputModelResp object	Definition: Output information of a custom training job.
asset_model	AssetModelResp object	Definition: Information about the model created using a custom training job.

**Table 6** Resource
Parameter	Type	Description
policy	String	Definition: Resource flavor mode of a training job. Range: regular: standard mode
flavor_id	String	Definition: Resource flavor ID of a training job. flavor_id cannot be specified for CPU-based dedicated resource pools. Range: The options for GPU- or Ascend-based dedicated resource pools are as follows: modelarts.pool.visual.xlarge (1 PU) modelarts.pool.visual.2xlarge (2 PUs) modelarts.pool.visual.4xlarge (4 PUs) modelarts.pool.visual.8xlarge (8 PUs)
flavor_name	String	Definition: Read-only flavor name returned by ModelArts when flavor_id is used. Range: N/A
node_count	Integer	Definition: Number of resource replicas selected for a training job. Range: greater than or equal to 1
pool_id	String	Definition: ID of the resource pool selected for a training job. Range: N/A
pool_group_id	String	Definition: ID of the resource pool federation selected for a training job. Range: N/A
flavor_detail	FlavorDetail object	Definition: Flavor details of a training job or algorithm. This parameter is available only for public resource pools.
main_container_allocated_resources	MainContainerAllocatedResources object	Definition: Resource specifications actually obtained by the training container of a training job.
main_container_customized_flavor	MainContainerCustomizedFlavor object	Definition: Custom flavor of a training job.

**Table 7** FlavorDetail
Parameter	Type	Description
flavor_type	String	Definition: Resource flavor type. Range: CPU: CPU resource specifications GPU: GPU resource specifications Ascend: NPU resource specifications
billing	BillingInfo object	Definition: Billing information of a resource flavor.
flavor_info	FlavorInfo object	Definition: Resource flavor details.

**Table 8** BillingInfo
Parameter	Type	Description
code	String	Definition: Billing code. Range: N/A
unit_num	Integer	Definition: Billing unit. Range: N/A

**Table 9** FlavorInfo
Parameter	Type	Description
max_num	Integer	Definition: Maximum number of nodes that can be selected. The value 1 indicates that the distributed mode is not supported. Range: N/A
cpu	Cpu object	Definition: CPU specifications.
gpu	Gpu object	Definition: GPU specifications.
npu	Npu object	Definition: Ascend specifications.
memory	Memory object	Definition: Memory information.
disk	Disk object	Definition: Disk information.

**Table 10** Cpu
Parameter	Type	Description
arch	String	Definition: CPU architecture. Range: N/A
core_num	Integer	Definition: Number of cores. Range: N/A

**Table 11** Gpu
Parameter	Type	Description
unit_num	Integer	Definition: Number of GPUs. Range: N/A
product_name	String	Definition: Product name. Range: N/A
memory	String	Definition: Memory size, in GB. Range: N/A

**Table 12** Npu
Parameter	Type	Description
unit_num	String	Definition: Number of NPUs. Range: N/A
product_name	String	Definition: Product name. Range: N/A
memory	String	Definition: Memory. Range: N/A

**Table 13** Memory
Parameter	Type	Description
size	Integer	Definition: Memory size. Range: N/A
unit	String	Definition: Number of memory units. Range: N/A

**Table 14** Disk
Parameter	Type	Description
size	String	Definition: Disk size. Range: N/A
unit	String	Definition: Unit of the disk size. Generally, the unit is GB. Range: N/A

**Table 15** MainContainerAllocatedResources
Parameter	Type	Description
cpu_arch	String	Definition: CPU architecture. Range: N/A
cpu_core_num	Float	Definition: Number of cores. Range: N/A
mem_size	Float	Definition: Memory information. Range: N/A
accelerator_num	Float	Definition: Number of accelerator cards. Range: N/A
accelerator_type	String	Definition: Type of accelerator cards. For example, ascend-d910b and ascend-snt9c. Range: N/A

**Table 16** MainContainerCustomizedFlavor
Parameter	Type	Description
cpu_core_num	Float	Definition: Number of CPU cores. Range: greater than 0
mem_size	Float	Definition: Memory size, in GB. Range: greater than 0
accelerator_num	Float	Definition: Number of accelerator cards. Range: greater than or equal to 0

**Table 17** JobVolumeResp
Parameter	Type	Description
nfs	NfsResp object	Definition: Volumes attached in NFS mode.

**Table 18** NfsResp
Parameter	Type	Description
nfs_server_path	String	Definition: NFS server path, for example, 10.10.10.10:/example/path. Range: N/A
local_path	String	Definition: Path for attaching volumes to the training container, for example, /example/path. Range: N/A
read_only	Boolean	Definition: Specifies whether the disks attached to the container in NFS mode are read-only. Range: true: read only false: non-read-only

**Table 19** LogExportPathResp
Parameter	Type	Description
obs_url	String	Definition: OBS path for storing training job logs, for example, obs://example/path. Range: N/A
host_path	String	Definition: Path of the host where training job logs are stored, for example, /example/path. Range: N/A

**Table 20** SchedulePolicyResp
Parameter	Type	Description
required_affinity	RequiredAffinityResp object	Definition: Affinity requirements of a training job.
priority	Integer	Definition: Priority of a training job. Range: 0 to 3
preemptible	Boolean	Definition: Whether the resource can be preempted. Range: true: The resource can be preempted. false: The resource cannot be preempted.

**Table 21** RequiredAffinityResp
Parameter	Type	Description
affinity_type	String	Definition: Affinity scheduling policy. Range: cabinet: strong cabinet scheduling hyperinstance: supernode affinity scheduling
job_level	String	Definition: Overall network topology constraint of a job. This parameter is only valid when affinity_type is set to networkTopology. The system schedules all tasks of the job to the node group at the level specified by job_level or lower. When you deliver a training job to a supernode resource pool, if the overall network topology constraint of the job is not set, the system assigns the value cluster by default. Range cluster: resource pool hyperinstanceGroup: supernode
affinity_group_size	Integer	Definition: Size of an affinity group. Range: N/A
affinity_group_level	String	Definition: Network topology constraint of an affinity group. This parameter is only valid when affinity_type is set to networkTopology. The system schedules the affinity group consisting of affinity_group_size tasks to a node group whose level is not higher than affinity_group_level. When you deliver a training job to the supernode resource pool, if the network topology constraints of the affinity group are not set, the system sets the value to hyperinstanceGroup by default. Range hyperinstance: supernode slice: cabinet

**Table 22** CustomMetrics
Parameter	Type	Description
exec	Exec object	Definition: Metrics are collected in CLI mode.
http_get	HttpGet object	Definition: Metrics are collected in HTTP mode.

**Table 23** Exec
Parameter	Type	Description
command	Array of strings	Definition: Metrics are collected in CLI mode.

**Table 24** HttpGet
Parameter	Type	Description
path	String	Definition: URL for obtaining metrics through HTTP. Range: N/A
port	Integer	Definition: Port for obtaining metrics through HTTP. Range: N/A

**Table 25** OutputModelResp
Parameter	Type	Description
obs	ObsModelResp object	Definition: OBS output information stored by a custom training job.

**Table 26** ObsModelResp
Parameter	Type	Description
obs_path	String	Definition: OBS path for storing custom training jobs, for example, obs://example/path. Range: N/A
local_path	String	Definition: Path of the host where custom training jobs are stored, for example, /example/path. Range: N/A

**Table 27** AssetModelResp
Parameter	Type	Description
id	String	Definition: Model ID. Range: N/A
name	String	Definition: Model name. Range: N/A
code	String	Definition: model code. Range: N/A
version	String	Definition: Model version. Range: N/A
location	String	Definition: Model address. Range: N/A
desc	String	Definition: Model description. Range: N/A
series	String	Definition: Model brand. Range: N/A
type	String	Definition: Model type. Range: N/A

**Table 28** DatasetConfig
Parameter	Type	Description
dataset_name	String	Training dataset name, which can be obtained from the name response body in the dataset list API.
dataset_source	String	Source of the used dataset. The value can be datamng (data engineering), OBS, or DB (database).
dataset_id	String	Training dataset ID, which can be obtained from the dataset_id response body in the dataset list API.
split_ratio	Integer	Splitting ratio of the training and validation datasets. This parameter is used when the model supports the validation dataset and the validation dataset is from the selected training dataset. The value is greater than or equal to 1 and less than or equal to 50.
used_step	String	Phase in which the dataset is used. The value can be train (training), eval (verification), or test.
dataset_proportion	Integer	Dataset combination, indicating the ratio of the dataset used for training.

**Table 29** Status
Parameter	Type	Description
phase	String	Definition: Level-1 status of a training job. Range: Creating: The job is being created. Pending: The job is pending. Running: The job is running. Failed: The job failed to run. Completed: The job is complete. Terminating: The job is being stopped. Terminated: The job has been stopped. Abnormal: The job is abnormal.
secondary_phase	String	Definition: Level-2 status of a training job. The values are internal detailed statuses and may be added, changed, or deleted. Dependency on the status is not recommended. Range: Creating: The job is being created. Queuing: The job is queuing. Running: The job is running. Failed: The job failed to run. Completed: The job is complete. Terminating: The job is being stopped. Terminated: The job has been stopped. CreateFailed: The job fails to be created. TerminatedFailed: The job fails to be stopped. Unknown: The job is in an unknown state. Lost: The job is abnormal.
duration	Long	Definition: Running duration of a training job, in ms. Range: N/A
node_count_metrics	Array<Array<Integer>>	Definition: number of nodes during training job running. Each inner array indicates a [time point, number of running nodes] 2-tuple, which records the number of nodes running at a specific time point.
tasks	Array of strings	Definition: Training job subtask name.
start_time	Long	Definition: training job start time. The value is a Unix timestamp, in milliseconds. Range: N/A
task_statuses	Array of TaskStatuses objects	Definition: Status of the first failed subtask of a training job.
running_records	Array of RunningRecord objects	Definition: Running and fault recovery records of a training job.
retention_time	Integer	Definition: Retention period of a job. reserved_time is set during training job creation. Range: N/A Default value: N/A
task_ips	Array of TaskIP objects	Definition: IP addresses of tasks in a training job. host_ips, and only the records that match the specified IP addresses are returned. Range: N/A host_ips is not passed, no value is returned.

**Table 30** TaskStatuses
Parameter	Type	Description
task	String	Definition: Training job subtask name. Range: N/A
exit_code	Integer	Definition: Exit code of a training job subtask. Range: N/A
message	String	Definition: Error message of a training job subtask. Range: N/A

**Table 31** RunningRecord
Parameter	Type	Description
start_at	Long	Definition: Unix timestamp of the start time in the current running record, in seconds. Range: N/A
end_at	Long	Definition: Unix timestamp of the end time in the current running record, in seconds. Range: N/A
xpu_start_at	Long	Definition: Unix timestamp of the accelerator card startup time in the current running record, in seconds. Range: N/A
start_type	String	Definition: Startup mode of the current execution. Range init_or_rescheduled: This startup is the first running after scheduling, including the first startup and the running after scheduling recovery. restarted: This startup is not the first running after scheduling but the running after a process restart.
end_reason	String	Definition: Reason why the running ends. Range: N/A
end_related_task	String	Definition: ID of the task worker (for example, worker-0) that ends the running. Range: N/A
end_recover	String	Definition: Fault tolerance policy adopted when the execution ends abnormally. Range npu_proc_restart: NPU in-place hot recovery proc_restart: in-place process recovery npu_step_retry: step recomputation pod_reschedule: pod-level rescheduling job_reschedule: job-level rescheduling job_reschedule_with_taint: isolated job-level rescheduling
end_recover_before_downgrade	String	Definition: There is a downgrade relationship between policies. If a policy fails to be executed, it will be downgraded to another specified policy. end_recover_before_downgrade indicates the tolerance policy used before end_recover is downgraded. Range: same as that of end_recover.
recover_records	Array of RecoverRecord objects	Definition: details about all fault tolerance policies adopted when the execution ends abnormally.

**Table 32** RecoverRecord
Parameter	Type	Description
recover_start_at	Long	Definition: Unix timestamp of the start time of the fault tolerance policy, in seconds. The timestamp is also the fault occurrence time. Range: N/A
recover_end_at	Long	Definition: Unix timestamp of the end time of the fault tolerance policy, in seconds. Range: N/A
recover	String	Definition: Fault tolerance policy. Range npu_step_retry: step recomputation npu_proc_restart: NPU in-place hot recovery proc_restart: in-place process recovery pod_reschedule: pod-level rescheduling job_reschedule: job-level rescheduling job_reschedule_with_taint: isolated job-level rescheduling
fault_scenario	String	Definition: Fault scenario. Range chip_fault: chip fault node_fault: node fault job_failed: job exit upon a failure job_hanged: job suspension job_subhealth: job subhealth error_in_log: log exception
reason	String	Definition: Fault cause. Range: N/A
related_task	String	Definition: ID of the task worker (for example, worker-0) that ends the running. Range: N/A
recover_result	String	Definition: Fault recovery result. Range recovering success failed downgrade: policy downgrade terminated: The policy is terminated. quotaExceeded: The number of policy executions exceeds the limit.

**Table 33** TaskIP
Parameter	Type	Description
task	String	Task name, for example, worker-0.
ip	String	Task or pod IP address.
host_ip	String	Host IP address.
schedule_count	Integer	Number of times the current task is scheduled. The default value is 1. The value increases in scenarios such as rescheduling and preemption.

**Table 34** ContinueTask
Parameter	Type	Description
checkpoint_id	String	Breakpoint ID
source_model_id	String	ID of the model of a resumable training task
source_model_name	String	Name of the model of a resumable training task
epoch	Integer	Number of rounds.
steps	Integer	Number of steps.
is_best	Boolean	Whether the value is optimal
skipped_steps	Integer	Number of steps to be skipped. 0 indicates that no step is skipped.

Status code: 400

**Table 35** Response body parameters
Parameter	Type	Description
error_msg	String	Error message
error_code	String	Error code
error_solution	String	Solution

Example Requests

The following shows how to query a training job whose UUID is 3faf5c03-aaa1-4cbe-879d-24b05d997347.

GET https://{endpoint}/v2/{project_id}/training-jobs/3faf5c03-aaa1-4cbe-879d-24b05d997347/ft-detail

Example Responses

Status code: 200

{
  "metadata" : {
    "id" : "3faf5c03-aaa1-4cbe-879d-24b05d997347",
    "name" : "trainjob--py14_mem06-108",
    "description" : "",
    "create_time" : 1636447346315,
    "workspace_id" : "0",
    "user_name" : ""
  },
  "status" : {
    "phase" : "Abnormal",
    "secondary_phase" : "CreateFailed",
    "duration" : 0,
    "start_time" : 0,
    "node_count_metrics" : [ [ 1636447746000, 0 ], [ 1636447755000, 0 ], [ 1636447756000, 0 ] ],
    "tasks" : [ "worker-0" ],
    "running_records" : [ {
      "start_at" : 1701327093,
      "end_at" : 1701322341,
      "start_type" : "init_or_rescheduled",
      "end_recover" : "job_reschedule",
      "end_reason" : "exit with 127",
      "end_related_task" : "worker-2",
      "end_recover_before_downgrade" : "npu_proc_restart"
    }, {
      "start_at" : 1701323345,
      "end_at" : 1701325432,
      "start_type" : "init_or_rescheduled",
      "end_reason" : "job completed"
    } ]
  },
  "spec" : {
    "resource" : {
      "flavor_id" : "modelarts.vm.pnt1.large.eco",
      "node_count" : 1,
      "flavor_detail" : {
        "flavor_type" : "GPU",
        "billing" : {
          "code" : "modelarts.vm.gpu.pnt1.eco",
          "unit_num" : 1
        },
        "flavor_info" : {
          "cpu" : {
            "arch" : "x86",
            "core_num" : 8
          },
          "gpu" : {
            "unit_num" : 1,
            "memory" : "8GB"
          },
          "memory" : {
            "size" : 64,
            "unit" : "GB"
          }
        }
      },
      "main_container_allocated_resources" : {
        "cpu_arch" : "x86",
        "cpu_core_num" : 5,
        "mem_size" : 44,
        "accelerator_num" : 1,
        "accelerator_type" : "nvidia-v100-pcie32"
      }
    },
    "custom_metrics" : [ {
      "exec" : {
        "command" : [ "cat", "/a/b/c.prom" ]
      }
    }, {
      "http_get" : {
        "path" : "/raw_text",
        "port" : 10001
      }
    } ]
  },
  "model_asset_id" : "14f39822-d31a-4ef1-b990-b606fe99496b"
}

Status code: 400

Format of the body for a common error response. The following shows the returned information when a training job with ID 3f5d6706-7b67-408d-8ba0-ec08048c45ee is not found.

{
  "error_msg" : "Job not found.",
  "error_code" : "ModelArts.2755",
  "error_solution" : "Check whether the training job in the request is valid."
}

Status Codes

Status Code	Description
200	ok
400	Format of the body for a common error response. The following shows the returned information when a training job with ID 3f5d6706-7b67-408d-8ba0-ec08048c45ee is not found.