Instructions to use Vionex-digital/Ar-CXR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Vionex-digital/Ar-CXR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Vionex-digital/Ar-CXR", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Vionex-digital/Ar-CXR", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Vionex-digital/Ar-CXR with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Vionex-digital/Ar-CXR" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Vionex-digital/Ar-CXR", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Vionex-digital/Ar-CXR
- SGLang
How to use Vionex-digital/Ar-CXR 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 "Vionex-digital/Ar-CXR" \ --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": "Vionex-digital/Ar-CXR", "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 "Vionex-digital/Ar-CXR" \ --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": "Vionex-digital/Ar-CXR", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Vionex-digital/Ar-CXR with Docker Model Runner:
docker model run hf.co/Vionex-digital/Ar-CXR
Download LICENSE.md from Vionex-digital/Ar-CXR: direct link, hf CLI and curl.
- Browser
- Download file 1.69 kB
-
https://huggingface.co/Vionex-digital/Ar-CXR/resolve/main/LICENSE.md
- Command line
-
hf download hf://Vionex-digital/Ar-CXR/LICENSE.md
-
curl -L -o LICENSE.md https://huggingface.co/Vionex-digital/Ar-CXR/resolve/main/LICENSE.md
Ar-CXR License (composite, research-only)
Ar-CXR is not distributed under a single permissive license. It is a derivative of multiple components, each with its own terms. You must comply with all of the upstream licenses below, and with the most restrictive where they overlap.
1. Ar-CXR adapters & code (this repository)
The trained adapters (weights/), configuration, and code (modeling_ar_cxr.py,
export_to_hf.py) are released by Vionex Digital Solutions for non-commercial
academic research only, with no warranty. They are not a medical device and
must not be used for clinical decision-making. Redistribution must retain this
notice and the disclaimers in README.md and NOTICE.md.
2. Decoder base — Falcon-H1-7B-Instruct
tiiuae/Falcon-H1-7B-Instruct is governed by the TII Falcon-LLM License 2.0.
You must accept and comply with it to download and use the base decoder:
https://huggingface.co/tiiuae/Falcon-H1-7B-Instruct
3. Vision base — RAD-DINO-MAIRA-2
microsoft/rad-dino-maira-2 is governed by its Microsoft Research license
(research use). You must accept and comply with it to download and use the base
vision encoder:
https://huggingface.co/microsoft/rad-dino-maira-2
4. Training data — CheXpert-Plus
Ar-CXR was trained on derivatives of CheXpert-Plus, governed by the Stanford CheXpert-Plus Data Use Agreement. The translated training corpus is not redistributed in this repository. The CheXpert-Plus terms (non-commercial research, no re-identification, no redistribution of the imaging data) extend to models and derivatives trained on it.
By downloading or using Ar-CXR you agree to all of the above.