---
{}
---

[![CODE](https://img.shields.io/badge/GitHub-Repository-<COLOR>)](https://github.com/mbzuai-oryx/LLaVA-pp)

# LLaMA-3-V: Extending the Visual Capabilities of LLaVA with Meta-Llama-3-8B-Instruct

## Repository Overview

This repository features LLaVA v1.5 trained with the Meta-Llama-3-8B-Instruct LLM. This integration aims to leverage the strengths of both models to offer advanced vision-language understanding.

## Training Strategy
- **Pretraining:** Only Vision-to-Language projector is trained. The rest of the model is frozen.
- **Fine-tuning:** LLM is LoRA fine-tuned. Only the vision-backbone (CLIP) is kept frozen.
- **Note:** The repository contains merged weights.

## Key Components

- **Base Large Language Model (LLM):** [Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)
- **Base Large Multimodal Model (LMM):** [LLaVA-v1.5](https://github.com/haotian-liu/LLaVA)

## Training Data

- **Pretraining Dataset:** [LCS-558K](https://huggingface.co/datasets/liuhaotian/LLaVA-Pretrain)
- **Fine-tuning Dataset:** [LLaVA-Instruct-665K](https://huggingface.co/datasets/liuhaotian/LLaVA-Instruct-150K/blob/main/llava_v1_5_mix665k.json)

## Download It As

```
git lfs install
git clone https://huggingface.co/MBZUAI/LLaVA-Meta-Llama-3-8B-Instruct
```

---


## Contributions

Contributions are welcome! Please 🌟 our repository [LLaVA++](https://github.com/mbzuai-oryx/LLaVA-pp) if you find this model useful.

---