README.md

---
license: mit
datasets:
- liuhaotian/LLaVA-Pretrain
- liuhaotian/LLaVA-Instruct-150K
language:
- en
metrics:
- accuracy
- precision
- recall
- f1
base_model:
- apple/aimv2-large-patch14-224
- apple/OpenELM
pipeline_tag: image-text-to-text
tags:
- cpu
- nano
- small
- tiny
- llava
model_size: 0.6B parameters
---

**<center><span style="font-size:2em;">Tiny Llava 4 CPU 🐛</span></center>**

[![License](https://img.shields.io/badge/License-MIT-brightgreen.svg)](https://opensource.org/licenses/MIT) 
[![CPU](https://img.shields.io/badge/CPU-Supported-blue)](https://huggingface.co)
[![arXiv](https://img.shields.io/badge/arXiv-2402.14289-red)](https://arxiv.org/pdf/2402.14289)

---

### 🚀 **Model Overview**
`tiny-llava-open-elm-aimv2` is a lightweight image-text-to-text model that combines **[OpenELM](https://huggingface.co/apple/OpenELM)** as the LLM backbone and **[AIMv2-Large-Patch14-224](https://huggingface.co/apple/aimv2-large-patch14-224)** as the vision encoder. The model has been fine-tuned using **LoRA (Low-Rank Adaptation)** for efficient training. It was developed using the **[TinyLLaVA Factory](https://github.com/TinyLLaVA/TinyLLaVA_Factory)** codebase, which provides a modular framework for lightweight multi-modal models. 

The model is designed to run efficiently on **CPU**, making it ideal for resource-constrained environments. It is trained and evaluated on **POPE** and **TextVQA** benchmarks. The total model size is **0.6B parameters**.

---

### 📊 **Performance**

|                          Model Name                          | VQAv2 | GQA   | SQA   | TextVQA | MM-VET | POPE  | MME    | MMMU  |
|:-----------------------------------------------------------:|:-----:|:-----:|:-----:|:-------:|:------:|:-----:|:------:|:-----:|
| [LLaVA-1.5-7B](https://huggingface.co/llava-hf/llava-1.5-7b-hf) | 78.5  | 62.0  | 66.8  |   58.2  |  30.5  | 85.9  | 1510.7 |   -   |
| [bczhou/TinyLLaVA-3.1B](https://huggingface.co/bczhou/TinyLLaVA-3.1B) | 79.9  | 62.0  | 69.1  |   59.1  |  32.0  | 86.4  | 1464.9 |   -   |
| [tinyllava/TinyLLaVA-Gemma-SigLIP-2.4B](https://huggingface.co/tinyllava/TinyLLaVA-Gemma-SigLIP-2.4B) | 78.4  | 61.6  | 64.4  |   53.6  |  26.9  | 86.4  | 1339.0 |  31.7 |
| [tinyllava/TinyLLaVA-Phi-2-SigLIP-3.1B](https://huggingface.co/tinyllava/TinyLLaVA-Phi-2-SigLIP-3.1B) | 80.1  | 62.1  | 73.0  |   60.3  |  37.5  | 87.2  | 1466.4 |  38.4 |
| tiny-llava-open-elm-aimv2                                   |   -   |   -   |   -   |  39.68  |   -    | 83.93 |   -    |   -   |

---

### 🔗 **References**
- [OpenELM](https://huggingface.co/apple/OpenELM)
- [AIMv2-Large-Patch14-224](https://huggingface.co/apple/aimv2-large-patch14-224)
- [TinyLLaVA Factory](https://github.com/TinyLLaVA/TinyLLaVA_Factory)
- [LoRA Paper (arXiv:2402.14289)](https://arxiv.org/pdf/2402.14289)