> For the complete documentation index, see [llms.txt](https://docs.aisuru.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aisuru.com/en/advanced-features/fine-tuning/how-to-do-fine-tuning-with-aisuru.md).

# How to do fine-tuning easily with AIsuru

AIsuru makes it easy to prepare data for fine-tuning language models on various AI platforms. This feature helps you create custom models that are more efficient and cost-effective, tailored to your specific needs.

### Preparing the data

To start the fine-tuning process, you'll first need to export your data from AIsuru:

1. From the sidebar, select "Import/Export";
2. Click the "**Export JSONL**" button;
3. In the advanced settings, select "Include instructions" if you want to include the "Instructions" (Settings > AI > Instructions) in every message;
4. (Optional) Specify a date to export only content from a certain period;
5. Press the "Export" button to start downloading the JSONL file.

✅ Great! You now have a JSONL file ready to feed to the language model!

{% hint style="danger" %}
The generated JSONL file only contains textual [Content](/en/agent-training/contents.md): it doesn't contain images, videos, links, PDFs, [known facts](/en/extensions/deep-thinking/managing-memories.md), [dynamic intents](/en/advanced-features/integrations/dynamic-intents.md), [slots](/en/advanced-features/integrations/dynamic-intents/how-to-use-slots.md), [functions](/en/advanced-features/integrations/functions.md), or [MCP](/en/advanced-features/integrations/mcp/mcp-what-they-are-and-how-to-use-them.md).
{% endhint %}

### The fine-tuning process

Once you have the JSONL file, you can proceed with fine-tuning:

1. Go to **Microsoft Azure** or [**OpenAI**](https://platform.openai.com/finetune);
2. **Upload the JSONL file** you downloaded from AIsuru **in the "Training data" section**;
3. Don't upload any file in the "Validation data" section and configure the other settings: suffix, seed, batch size, learning rate multiplier, and number of epochs.

✅ Done! The process should complete within a few hours.

Fine-tuning with AIsuru lets you create custom AI models that perfectly reflect your needs, reducing costs and improving efficiency. Give it a try and see how it can transform your approach to AI!