Get fine-tuning job status and artifacts¶
Read current job state, training inputs, workflow associations, errors, and adapter URI. Continue querying according to these state rules. succeeded does not automatically establish actual weight training or a published inference service.
GET https://api.moi.matrixorigin.cn/v5/workspaces/{id}/finetune-jobs/{job_id}
Before you call this API¶
Set BASE_URL to the product API prefix https://api.moi.matrixorigin.cn/v5; use the corresponding product API URL for a dedicated deployment.
Prepare identity, training inputs, and environment. Example IDs and results are fixed illustrative values. Replace models, run_id, request_id, and compute instances before calling.
Request¶
curl --request GET \
"$BASE_URL/workspaces/$WORKSPACE_ID/finetune-jobs/$JOB_ID" \
--header "X-API-Key: $MOI_KEY" \
--header "X-Workspace-ID: $WORKSPACE_ID"
Parameter |
Location |
Type |
Required |
Description |
|---|---|---|---|---|
|
Path |
string |
Yes |
Target workspace ID, matching X-Workspace-ID. |
|
Headers |
string |
Yes |
Workspace accessible to the current identity in the same environment. |
|
Path |
string |
Yes |
Fine-tuning job_id returned by successful creation. |
Successful response¶
{
"code": "OK",
"msg": "OK",
"data": {
"job_id": "ftj-example",
"workspace_id": "workspace-example",
"request_id": "request-example",
"base_model": "example/base-model",
"status": "succeeded",
"train_run_ids": [
"train-run-1",
"train-run-2"
],
"holdout_run_ids": [
"holdout-run-1"
],
"recipe": {
"training_objective": "tool_selector"
},
"compute_resource_id": "resource-example",
"created_at": "2026-11-12T01:00:00Z",
"updated_at": "2026-11-12T01:00:00Z",
"mowl_task_id": "workflow-task-example",
"mowl_case_id": "workflow-case-example",
"started_at": "2026-11-12T01:00:00Z",
"adapter_uri": "s3://example-adapters/finetune/workspace-example/ftj-example/",
"finished_at": "2026-11-12T01:00:00Z"
}
}
Field |
Type |
Description |
|---|---|---|
|
string |
OK on success. |
|
string |
Response information. |
|
object |
Errors and next steps¶
Lifecycle and recovery distinguishes parameters, identity, conflicts, saved jobs, and actual training outcomes.
Continue querying¶
While pending/running, repeat the same complete request at deployment-approved intervals. Stop at succeeded, failed, or cancelled. For failures, inspect error_code/error_message. For success, inspect adapter_uri and the deployment execution mode. These APIs currently have no list, delete, retry, or publication endpoint. Creating a new job does not replace status verification.