Implementation:Googleapis Python genai Tunings Get

Knowledge Sources	googleapis/python-genai Google Gen AI Python SDK
Domains	Fine_Tuning, Operations
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete tool for retrieving tuning job status and monitoring progress provided by the google-genai tunings module.

Description

Tunings.get retrieves the current state of a tuning job by its resource name. It returns an updated TuningJob object with the current state (JOB_STATE_SUCCEEDED, JOB_STATE_FAILED, etc.), tuned_model (available when complete), training metrics, and timestamps. This is used in polling loops to monitor job progress.

Usage

Call client.tunings.get(name=job_name) periodically to check job status. The name parameter is the resource name from the TuningJob returned by tunings.tune. Check tuning_job.state for completion status.

Code Reference

Source Location

• Repository: googleapis/python-genai
• File: google/genai/tunings.py
• Lines: L1703

Signature

class Tunings:
    def get(
        self,
        *,
        name: str,
        config: Optional[types.GetTuningJobConfigOrDict] = None,
    ) -> types.TuningJob:
        """Gets a tuning job by name.

        Args:
            name: Tuning job resource name.
            config: Optional request configuration.
        """

Import

from google import genai

I/O Contract

Inputs

Name	Type	Required	Description
name	str	Yes	Tuning job resource name from TuningJob.name
config	Optional[GetTuningJobConfigOrDict]	No	Optional request config

Outputs

Name	Type	Description
TuningJob	types.TuningJob	Updated job with .state, .tuned_model, .create_time, .end_time

Usage Examples

Polling Loop

import time
from google import genai

client = genai.Client(vertexai=True, project="my-project", location="us-central1")

# After launching a job
job_name = tuning_job.name

while True:
    job = client.tunings.get(name=job_name)
    print(f"State: {job.state}")

    if job.state == "JOB_STATE_SUCCEEDED":
        print(f"Tuned model: {job.tuned_model.endpoint}")
        break
    elif job.state in ("JOB_STATE_FAILED", "JOB_STATE_CANCELLED"):
        print(f"Job ended: {job.state}")
        break

    time.sleep(60)  # Poll every 60 seconds

Use Tuned Model After Completion

# Once job succeeds, use the tuned model
job = client.tunings.get(name=job_name)
if job.state == "JOB_STATE_SUCCEEDED":
    response = client.models.generate_content(
        model=job.tuned_model.endpoint,
        contents="Classify: This product is amazing!"
    )
    print(response.text)

Related Pages

Implements Principle

• Principle:Googleapis_Python_genai_Job_Monitoring