Check Publish Model Readiness

  • Published on Nov 14, 2025
  • 1 minute(s) read
  • eA
 Prev Next

For cost and resource optimization, text clustering published pods that have not received requests for a certain period will timeout and become temporarily unavailable. This endpoint checks whether the published model is ready to serve inference requests.



Endpoint:

URL: {{baseUrl}}/sestekai-api/api/external/trainings/{modelId}/ready
HTTP Method: GET
Content-Type: none


The access token acquired for the training must be included in the request header.


Here's an example of a request:

curl --location '{{baseUrl}}/sestekai-api/api/external/trainings/:modelId/ready' \
--header 'Authorization: Bearer <token>'

Expected Response: 200 OK & Response Body: True/False

true  // The published model is ready and operational. Inference requests can be submitted.

false // The published model is unavailable (timed-out due to inactivity). Training must be initiated to restore the model to operational status.


If the response is false, a training request must be initiated to restore the model with the parameter forceTrain;

  • If forceTrain=false → The existing model will be redeployed without retraining.
  • If forceTrain=true → The model will be retrained with the updated data before deployment.