How to beginner · 4 min read

How to start fine-tuning job with OpenAI API

Quick answer

To start a fine-tuning job with the OpenAI API, first prepare and upload your training data in JSONL format, then call client.fine_tunes.create() with the uploaded file ID and desired base model. The API will return a fine-tune job ID to track progress and retrieve the fine-tuned model.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the OpenAI Python SDK and set your API key as an environment variable for secure authentication.

bash

pip install openai>=1.0

Step by step

Prepare your training data in JSONL format, upload it, then start the fine-tuning job by specifying the base model and training file ID.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Step 1: Upload your training data file (JSONL format)
with open("training_data.jsonl", "rb") as f:
    upload_response = client.files.create(
        file=f,
        purpose="fine-tune"
    )

training_file_id = upload_response.id
print(f"Uploaded training file ID: {training_file_id}")

# Step 2: Create fine-tuning job
fine_tune_response = client.fine_tunes.create(
    training_file=training_file_id,
    model="gpt-4o"
)

print(f"Fine-tuning job started with ID: {fine_tune_response.id}")

# Optional: Check fine-tune job status
status_response = client.fine_tunes.get(id=fine_tune_response.id)
print(f"Status: {status_response.status}")

output

Uploaded training file ID: file-abc123xyz
Fine-tuning job started with ID: ft-xyz789abc
Status: pending

Common variations

Use different base models like gpt-4o-mini or gpt-4o depending on your needs.
Use asynchronous calls or polling to track fine-tune job progress.
Include validation files by uploading a validation_file parameter.

Troubleshooting

If you get an error about file format, ensure your training data is valid JSONL with prompt and completion fields.
If the fine-tune job fails, check the status and events via client.fine_tunes.get() for detailed error messages.
Ensure your API key has permissions for fine-tuning.

✅

Key Takeaways

Upload your training data as a JSONL file with prompt and completion fields before starting fine-tuning.
Use client.fine_tunes.create() with the training file ID and base model to start the fine-tuning job.
Track fine-tuning status with client.fine_tunes.get() and handle errors by inspecting job events.
Choose the base model carefully based on your use case and resource constraints.
Always keep your API key secure in environment variables and never hardcode it.

Verified 2026-04 · gpt-4o, gpt-4o-mini

Verify ↗