Instructions to use andyc03/Qwen2-VL-PRISM-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use andyc03/Qwen2-VL-PRISM-SFT with Transformers:

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

pipe = pipeline("image-text-to-text", model="andyc03/Qwen2-VL-PRISM-SFT")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("andyc03/Qwen2-VL-PRISM-SFT")
model = AutoModelForMultimodalLM.from_pretrained("andyc03/Qwen2-VL-PRISM-SFT", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use andyc03/Qwen2-VL-PRISM-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "andyc03/Qwen2-VL-PRISM-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "andyc03/Qwen2-VL-PRISM-SFT",
		"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/andyc03/Qwen2-VL-PRISM-SFT

SGLang

How to use andyc03/Qwen2-VL-PRISM-SFT 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 "andyc03/Qwen2-VL-PRISM-SFT" \
    --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": "andyc03/Qwen2-VL-PRISM-SFT",
		"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 "andyc03/Qwen2-VL-PRISM-SFT" \
        --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": "andyc03/Qwen2-VL-PRISM-SFT",
		"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 andyc03/Qwen2-VL-PRISM-SFT with Docker Model Runner:
```
docker model run hf.co/andyc03/Qwen2-VL-PRISM-SFT
```

Add model card for PRISM (Qwen2-VL-PRISM-SFT)

by nielsr HF Staff - opened Aug 31, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+49

-0

nielsr

Aug 31, 2025

This PR adds a comprehensive model card for the Qwen2-VL-PRISM-SFT model, which is a core component of the PRISM framework for robust Vision-Language Model (VLM) alignment.

Key additions and improvements include:

Metadata: Added license: mit, library_name: transformers, and pipeline_tag: image-text-to-text. These tags enhance discoverability on the Hugging Face Hub and enable the "Use in Transformers" widget, providing automated code snippets for users.
Paper Link: Linked the model to its associated paper, PRISM: Robust VLM Alignment with Principled Reasoning for Integrated Safety in Multimodality.
GitHub Repository: Included a direct link to the official GitHub repository: https://github.com/SaFoLab-WISC/PRISM.
Abstract: Incorporated the paper's abstract to provide a concise overview of the model's purpose and methodology.
Model Details: Added information about the training datasets and model weights mentioned in the GitHub README.
Citation: Included the BibTeX entry for easy referencing of the paper.

These updates significantly improve the model's documentation, making it more informative and accessible for the community.

Add model card for PRISM (Qwen2-VL-PRISM-SFT)d9413d4f

andyc03 changed pull request status to merged Sep 1, 2025

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment