# Fine-tune with LoRA

### How to create a Fine-tuning job with LoRA? <a href="#contentify_0" id="contentify_0"></a>

<figure><img src="/files/q853xgkvhF6od4OIZfHP" alt=""><figcaption></figcaption></figure>

To fine-tune a model with LoRA, please follow the instructions below:

**Notes**

* You must log in before starting a fine-tune job.
* Ensure you have enough balance (credit).
* At least one base model must be available for fine-tuning.

**Steps**

1. **Go to the Fine-tuning Jobs page** and click **+ Fine-tune model**.
2. In the pop-up, enter the **Name** of your fine-tuning job.
   * Validation: Required, max 100 characters, supports Unicode letters, digits, `-`, `_`, `.`

<figure><img src="/files/JwhIUgskJ61qLKflD8Uq" alt=""><figcaption></figcaption></figure>

3. **Select a Base model** from the dropdown list.
   * Examples: `gemma-3-27b-it`, `Qwen3-4B-Instruct-2507`, `Llama-3.3-70B-Instruct`
4. **Select dataset format** from the dropdown list: Alpaca/ ShareGPT/ ShareGPT\_Image
5. **Upload your Training data file**.
   * Supported formats: CSV, JSON, JSONL, ZIP, Parquet (<100MB).
6. *(Optional)* **Upload Validation data**.
7. *(Optional)* **Configure hyperparameters**:
   * **Learning rate:** Float, `1e-6 → 1e-4` (e.g., `0.00001`)
   * **Number of epochs:** Integer `1–20` (default = `5`)
8. Click **Create** to start the fine-tuning job.

   * The job will appear in the table with status **Running**.

   > **Note:** Fine-tuning with LoRA usually takes only a few minutes.

***

### How to manage Fine-tuning jobs? <a href="#contentify_1" id="contentify_1"></a>

On the **Fine-tuning Jobs** page, you can:

* **View detail:** Open the pipeline detail in AI Studio.
* **Deploy model:** Once training is completed, deploy the LoRA model.
* **Cancel job:** Cancel a running job (requires confirmation).
* **Delete job:** Permanently delete a job (requires confirmation).

**Status badges**

* **Running** (yellow)
* **Succeeded** (green)
* **Failed** (red)
* **Canceled** (gray)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://ai-docs.fptcloud.com/fpt-ai-inference/fpt-ai-inference/tutorials/fine-tune-with-lora.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.