README.md

---
license: apache-2.0
pipeline_tag: image-text-to-text
library_name: transformers
paper: https://arxiv.org/abs/2409.03277
---


<p align="center">
    <b><font size="6">ChartMoE</font></b> 
<p>
<p align="center">
    <b><font size="4">ICLR2025 Oral </font></b>
<p>

<div align="center">
<div style="display: inline-block; margin-right: 30px;">
  
  [![arXiv](https://img.shields.io/badge/ArXiv-Prepint-red)](https://arxiv.org/abs/2409.03277)
</div>
<div style="display: inline-block; margin-right: 30px;">
  
  [![Project Page](https://img.shields.io/badge/Project-Page-brightgreen)](https://chartmoe.github.io/)
</div>
<div style="display: inline-block; margin-right: 30px;">
  
  [![Github Repo](https://img.shields.io/badge/Github-Repo-blue)](https://github.com/IDEA-FinAI/ChartMoE)
</div>
<div style="display: inline-block; margin-right: 30px;">
  
  [![Hugging Face Dataset](https://img.shields.io/badge/Hugging%20Face-Dataset-8A2BE2)](https://huggingface.co/datasets/Coobiw/ChartMoE-Data)
</div>
</div>


![](teaser.png)

**ChartMoE** is a multimodal large language model with Mixture-of-Expert connector, based on [InternLM-XComposer2](https://github.com/InternLM/InternLM-XComposer/tree/main/InternLM-XComposer-2.0) for advanced chart 1)understanding, 2)replot, 3)editing, 4)highlighting and 5)transformation. 


## Import from Transformers
To load the ChartMoE model using Transformers, use the following code:
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
ckpt_path = "IDEA-FinAI/chartmoe"
tokenizer = AutoTokenizer.from_pretrained(ckpt_path, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(ckpt_path, trust_remote_code=True).half().cuda().eval()
```

## Quickstart & Gradio Demo
We provide a simple example and a gradio webui demo to show how to use ChartMoE. Please refer to [https://github.com/IDEA-FinAI/ChartMoE](https://github.com/IDEA-FinAI/ChartMoE).


## Open Source License
The code is licensed under Apache-2.0.