Instructions to use byrLLCC/ChestX-Reasoner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use byrLLCC/ChestX-Reasoner with Transformers:

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

pipe = pipeline("image-text-to-text", model="byrLLCC/ChestX-Reasoner")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("byrLLCC/ChestX-Reasoner")
model = AutoModelForImageTextToText.from_pretrained("byrLLCC/ChestX-Reasoner")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use byrLLCC/ChestX-Reasoner with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "byrLLCC/ChestX-Reasoner"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "byrLLCC/ChestX-Reasoner",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/byrLLCC/ChestX-Reasoner

SGLang

How to use byrLLCC/ChestX-Reasoner 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 "byrLLCC/ChestX-Reasoner" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "byrLLCC/ChestX-Reasoner",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "byrLLCC/ChestX-Reasoner" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "byrLLCC/ChestX-Reasoner",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use byrLLCC/ChestX-Reasoner with Docker Model Runner:
```
docker model run hf.co/byrLLCC/ChestX-Reasoner
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

ChestX-Reasoner

Access to this model is restricted and requires manual approval.

This repository contains the model weights for ChestX-Reasoner, a model developed for chest X-ray reasoning and report-related research.

Access Policy

This model is released as a gated model with manual review.

To request access, please:

Click Request access on Hugging Face.
Send an email to liangcheng@sjtu.edu.cn
Include proof that you have authorized access to the MIMIC database (or the relevant MIMIC resource used in your research setting).

Access requests will only be approved after manual verification.

Required Materials for Access Request

When emailing liangcheng@sjtu.edu.cn, please include:

Your full name
Your institution / affiliation
Your Hugging Face username
Your intended use of the model
Proof of authorized access to MIMIC

Examples of acceptable proof may include:

A screenshot showing your authorized access status
A confirmation email or credential page
Other reasonable documentation demonstrating that you have valid MIMIC access rights

If your request does not include sufficient proof of authorized access, it may be declined.

Important Ethical Notice

This model is for research purposes only.

Because the training source involves access-controlled medical data, special care is required regarding:

privacy
memorization risk
redistribution risk
secondary release of potentially sensitive content
compliance with source dataset governance

If you are unsure whether your intended use is permissible, please contact liangcheng@sjtu.edu.cn before using the model.

Liang Cheng
Email: liangcheng@sjtu.edu.cn

Downloads last month: -

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for byrLLCC/ChestX-Reasoner

Base model

Qwen/Qwen2.5-VL-7B-Instruct

Finetuned

(1064)

this model