Image-Text-to-Text
Transformers
ONNX
Safetensors
English
medical
chest-xray
radiology
clip
blip
multimodal
cpu
Instructions to use GAD-Research-Lab/MedicalAI-Light-Weight with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use GAD-Research-Lab/MedicalAI-Light-Weight with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="GAD-Research-Lab/MedicalAI-Light-Weight")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("GAD-Research-Lab/MedicalAI-Light-Weight", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use GAD-Research-Lab/MedicalAI-Light-Weight with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "GAD-Research-Lab/MedicalAI-Light-Weight" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "GAD-Research-Lab/MedicalAI-Light-Weight", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/GAD-Research-Lab/MedicalAI-Light-Weight
- SGLang
How to use GAD-Research-Lab/MedicalAI-Light-Weight 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 "GAD-Research-Lab/MedicalAI-Light-Weight" \ --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": "GAD-Research-Lab/MedicalAI-Light-Weight", "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 "GAD-Research-Lab/MedicalAI-Light-Weight" \ --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": "GAD-Research-Lab/MedicalAI-Light-Weight", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use GAD-Research-Lab/MedicalAI-Light-Weight with Docker Model Runner:
docker model run hf.co/GAD-Research-Lab/MedicalAI-Light-Weight
| """ | |
| Build a standalone .exe for Windows using PyInstaller. | |
| No Python installation needed on the target machine. | |
| Usage: | |
| python build_exe.py # build web_ui.exe | |
| python build_exe.py --cli # build run.exe (CLI) | |
| python build_exe.py --all # build both | |
| Requires: pip install pyinstaller | |
| """ | |
| import argparse | |
| import os | |
| import shutil | |
| import subprocess | |
| import sys | |
| DIST_DIR = "./dist" | |
| def build_exe(script, name=None): | |
| if name is None: | |
| name = os.path.splitext(os.path.basename(script))[0] | |
| print(f"[cyan]Building {name}.exe from {script}...[/cyan]") | |
| cmd = [ | |
| sys.executable, "-m", "PyInstaller", | |
| "--onefile", | |
| "--console", | |
| "--name", name, | |
| "--distpath", DIST_DIR, | |
| "--workpath", "./build", | |
| "--specpath", "./build", | |
| "--add-data", "config.json;.", | |
| script, | |
| ] | |
| subprocess.check_call(cmd) | |
| exe_path = os.path.join(DIST_DIR, f"{name}.exe") | |
| if os.path.exists(exe_path): | |
| size_mb = os.path.getsize(exe_path) / 1024 / 1024 | |
| print(f"[green] Created: {exe_path} ({size_mb:.0f} MB)[/green]") | |
| else: | |
| print(f"[red] Failed to create {name}.exe[/red]") | |
| # Clean up build artifacts | |
| for d in ["./build", "*.spec"]: | |
| try: | |
| if os.path.isdir(d): | |
| shutil.rmtree(d) | |
| except Exception: | |
| pass | |
| for f in os.listdir("."): | |
| if f.endswith(".spec"): | |
| os.remove(f) | |
| def main(): | |
| parser = argparse.ArgumentParser(description="Build standalone executable") | |
| parser.add_argument("--cli", action="store_true", help="Build CLI executable") | |
| parser.add_argument("--all", action="store_true", help="Build all executables") | |
| args = parser.parse_args() | |
| try: | |
| import PyInstaller # noqa: F401 | |
| except ImportError: | |
| print("PyInstaller is required. Install with: pip install pyinstaller") | |
| sys.exit(1) | |
| os.makedirs(DIST_DIR, exist_ok=True) | |
| if args.all: | |
| build_exe("web_ui.py") | |
| build_exe("run.py") | |
| build_exe("batch_predict.py") | |
| elif args.cli: | |
| build_exe("run.py") | |
| else: | |
| build_exe("web_ui.py") | |
| print(f"\n[green]Done. Executables in ./{DIST_DIR}/[/green]") | |
| print("[yellow]Note: The .exe still needs model files (checkpoints/).[/yellow]") | |
| print("[yellow]Copy the entire project folder to the target machine.[/yellow]") | |
| if __name__ == "__main__": | |
| main() | |