Instructions to use ChartFoundation/ECD_Finetuned_MLLMs with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ChartFoundation/ECD_Finetuned_MLLMs with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="ChartFoundation/ECD_Finetuned_MLLMs")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ChartFoundation/ECD_Finetuned_MLLMs", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ChartFoundation/ECD_Finetuned_MLLMs with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ChartFoundation/ECD_Finetuned_MLLMs"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChartFoundation/ECD_Finetuned_MLLMs",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ChartFoundation/ECD_Finetuned_MLLMs

SGLang

How to use ChartFoundation/ECD_Finetuned_MLLMs with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ChartFoundation/ECD_Finetuned_MLLMs" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChartFoundation/ECD_Finetuned_MLLMs",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ChartFoundation/ECD_Finetuned_MLLMs" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChartFoundation/ECD_Finetuned_MLLMs",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ChartFoundation/ECD_Finetuned_MLLMs with Docker Model Runner:
```
docker model run hf.co/ChartFoundation/ECD_Finetuned_MLLMs
```

nielsr HF Staff commited on Aug 11, 2025

Commit

ff581dd

verified ·

1 Parent(s): 9e04a6a

Improve model card: add correct pipeline tag, library_name, and relevant links

Browse files

This PR improves the model card for the model presented in [Effective Training Data Synthesis for Improving MLLM Chart Understanding](https://huggingface.co/papers/2508.06492).

Specifically, it:
- Corrects the `pipeline_tag` to `image-text-to-text` for better discoverability on the Hugging Face Hub.
- Adds `library_name: transformers` as the model is compatible with the 🤩 Transformers library, making it easier for users to identify how to use it.
- Updates the introductory sentence to include a direct link to the Hugging Face paper page and the GitHub repository for easier access to the code.

Please review and merge this PR if everything looks good.

Files changed (1) hide show

README.md +7 -4

README.md CHANGED Viewed

@@ -1,14 +1,17 @@
 ---
-license: mit
-metrics:
-- accuracy
 base_model:
 - llava-hf/llama3-llava-next-8b-hf
 - openbmb/MiniCPM-V-2_6
 - microsoft/Phi-3-vision-128k-instruct
 - Qwen/Qwen2.5-VL-7B-Instruct
 ---
-**The following model are obtained via supervised fine-tuning (SFT) using the ECD-10k-Images dataset (URL: https://huggingface.co/datasets/ChartFoundation/ECD-10k-Images) proposed in our ICCV 2025 paper, "Effective Training Data Synthesis for Improving MLLM Chart Understanding."**
 **ECD Dataset Overview**:
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6666a432b78b8b6b34a816e9/kKQmAbuLKB7zOmOVegaOe.png)

 ---
 base_model:
 - llava-hf/llama3-llava-next-8b-hf
 - openbmb/MiniCPM-V-2_6
 - microsoft/Phi-3-vision-128k-instruct
 - Qwen/Qwen2.5-VL-7B-Instruct
+license: mit
+metrics:
+- accuracy
+pipeline_tag: image-text-to-text
+library_name: transformers
 ---
+**The following models are obtained via supervised fine-tuning (SFT) using the ECD-10k-Images dataset ([URL](https://huggingface.co/datasets/ChartFoundation/ECD-10k-Images)) proposed in our ICCV 2025 paper, "[Effective Training Data Synthesis for Improving MLLM Chart Understanding](https://huggingface.co/papers/2508.06492)" ([Code](https://github.com/yuweiyang-anu/ECD)).**
 **ECD Dataset Overview**:
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6666a432b78b8b6b34a816e9/kKQmAbuLKB7zOmOVegaOe.png)