Cancel a nonterminal fine-tuning job¶
If a real workflow binding exists, cancellation is requested first, then cancelled is persisted. Terminal jobs return 409. This does not delete history, recover all files, or immediately stop every training subprocess.
POST https://api.moi.matrixorigin.cn/v5/workspaces/{id}/finetune-jobs/{job_id}/cancel
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 POST \
"$BASE_URL/workspaces/$WORKSPACE_ID/finetune-jobs/$JOB_ID/cancel" \
--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": "cancelled",
"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",
"error_code": "cancelled",
"error_message": "cancelled by user",
"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.