---
title: "Comment by lin lin on \"Fine-Tune Llama 3.2 Vision-Language Model on Custom Datasets\" | Hashnode"
description: "Thaks for your excellent blog! I have some questions that can I make this fintuned model based on the unscloth frame to get the embeddings?That means can I just"
url: "https://hashnode.com/posts/fine-tune-llama-32-vision-language-model-on-custom-datasets/671268c74754043681a0dfeb/comment/6768f24523807df9c9cad328"
---

[Discussion](https://hashnode.com/posts/fine-tune-llama-32-vision-language-model-on-custom-datasets/671268c74754043681a0dfeb)

Comment [Fine-Tune Llama 3.2 Vision-Language Model on Custom Datasets](https://hashnode.com/posts/fine-tune-llama-32-vision-language-model-on-custom-datasets/671268c74754043681a0dfeb)

[lin lin](https://hashnode.com/@linlin0910)

[Dec 23, 2024](https://hashnode.com/posts/fine-tune-llama-32-vision-language-model-on-custom-datasets/671268c74754043681a0dfeb/comment/6768f24523807df9c9cad328)

Thaks for your excellent blog! I have some questions that can I make this fintuned model based on the unscloth frame to get the embeddings?That means can I just input a image, and I can get the image-text embedding from finetune llama?
