tunings.get

Retrieve information about a tuning job

Method

client.tunings.get(
    name: str,
    config: Optional[GetTuningJobConfig] = None
) -> TuningJob

Retrieves detailed information about a specific tuning job by its resource name.

name · string · required

The resource name of the tuning job. Format:

• Vertex AI: projects/{project}/locations/{location}/tuningJobs/{job_id}
• Gemini API: tunedModels/{model_id}

config · GetTuningJobConfig

Optional configuration for the request

Show Configuration Fields

http_options · HttpOptions

Custom HTTP options for the request

Response

name · string

The resource name of the tuning job

state · JobState

Current state of the tuning job:

• JOB_STATE_QUEUED
• JOB_STATE_PENDING
• JOB_STATE_RUNNING
• JOB_STATE_SUCCEEDED
• JOB_STATE_FAILED
• JOB_STATE_CANCELLED

create_time · string

ISO 8601 timestamp when the job was created

start_time · string

ISO 8601 timestamp when the job started

end_time · string

ISO 8601 timestamp when the job completed

update_time · string

ISO 8601 timestamp of the last update

base_model · string

The base model that is being tuned

tuned_model · TunedModel

Information about the resulting tuned model (available after completion)

Show TunedModel Fields

model · string

The resource name of the tuned model

endpoint · string

The endpoint to use for the tuned model

error · object

Error information if the job failed

description · string

Description of the tuning job (Vertex AI only)

tuning_data_stats · object

Statistics about the training data (Vertex AI only)

Usage

Check Job Status

from google import genai

client = genai.Client(api_key='your-api-key')

# Get tuning job details
tuning_job = client.tunings.get(
    name='projects/my-project/locations/us-central1/tuningJobs/123456'
)

print(f"Job: {tuning_job.name}")
print(f"State: {tuning_job.state}")
print(f"Created: {tuning_job.create_time}")

if tuning_job.state == 'JOB_STATE_SUCCEEDED':
    print(f"Tuned model: {tuning_job.tuned_model.model}")

Monitor Job Progress

import time
from google.genai import types

job_name = 'projects/my-project/locations/us-central1/tuningJobs/123456'

while True:
    tuning_job = client.tunings.get(name=job_name)
    
    print(f"Current state: {tuning_job.state}")
    
    # Check if job is complete
    if tuning_job.state in [
        types.JobState.JOB_STATE_SUCCEEDED,
        types.JobState.JOB_STATE_FAILED,
        types.JobState.JOB_STATE_CANCELLED
    ]:
        break
    
    # Wait before checking again
    time.sleep(60)

if tuning_job.state == types.JobState.JOB_STATE_SUCCEEDED:
    print(f"Success! Tuned model: {tuning_job.tuned_model.model}")
else:
    print(f"Job ended with state: {tuning_job.state}")
    if tuning_job.error:
        print(f"Error: {tuning_job.error}")

Handle Job Completion

from google.genai import types

tuning_job = client.tunings.get(name=job_name)

# Check job status and take action
if tuning_job.state == types.JobState.JOB_STATE_SUCCEEDED:
    # Use the tuned model
    model = tuning_job.tuned_model.model
    response = client.models.generate_content(
        model=model,
        contents='Test the tuned model'
    )
    print(response.text)
    
elif tuning_job.state == types.JobState.JOB_STATE_FAILED:
    print(f"Tuning failed: {tuning_job.error}")
    
elif tuning_job.state == types.JobState.JOB_STATE_RUNNING:
    elapsed = tuning_job.update_time - tuning_job.start_time
    print(f"Job still running, elapsed time: {elapsed}")

See Also

• tunings.tune - Create a new tuning job
• tunings.list - List all tuning jobs
• tunings.cancel - Cancel a running job