Instructions to use osunlp/UGround-V1-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use osunlp/UGround-V1-2B with Transformers:

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

pipe = pipeline("image-text-to-text", model="osunlp/UGround-V1-2B", 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?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("osunlp/UGround-V1-2B")
model = AutoModelForMultimodalLM.from_pretrained("osunlp/UGround-V1-2B", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use osunlp/UGround-V1-2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "osunlp/UGround-V1-2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "osunlp/UGround-V1-2B",
		"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/osunlp/UGround-V1-2B

SGLang

How to use osunlp/UGround-V1-2B 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 "osunlp/UGround-V1-2B" \
    --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": "osunlp/UGround-V1-2B",
		"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 "osunlp/UGround-V1-2B" \
        --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": "osunlp/UGround-V1-2B",
		"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 osunlp/UGround-V1-2B with Docker Model Runner:
```
docker model run hf.co/osunlp/UGround-V1-2B
```

[BUG] Trying to run vLLM inference crashes with an AttributeError

by ntoxeg - opened Jan 7, 2025

Discussion

ntoxeg

Jan 7, 2025

Trying to run using the vLLM command provided in the model card (vLLM v0.6.6) results in AttributeError: 'Qwen2VLProcessor' object has no attribute 'image_token'.

BoyuNLP

OSU NLP Group org Jan 7, 2025

the version I used:
vllm 0.6.5

Not pretty sure if that is due to the version of vllm used.

BoyuNLP

OSU NLP Group org Jan 7, 2025

Could you check if you can run the original Qwen2-VL-Instruct successfully?

BoyuNLP

OSU NLP Group org Jan 7, 2025

•

edited Jan 7, 2025

probably from the version of transformers. Did you follow the instructions in Qwen2-VL's doc?

The version in my env:
transformers 4.47.1

BoyuNLP changed discussion status to closed Jan 7, 2025

BoyuNLP changed discussion status to open Jan 7, 2025

BoyuNLP

OSU NLP Group org Jan 7, 2025

•

edited Jan 7, 2025

https://github.com/QwenLM/Qwen2-VL

Deployment
We recommend using vLLM for fast Qwen2-VL deployment and inference. You need to use vllm>=0.6.1 to enable Qwen2-VL support. You can also use our official docker image.

Installation
pip install git+https://github.com/huggingface/transformers@21fac7abba2a37fae86106f87fcf9974fd1e3830
pip install accelerate
pip install qwen-vl-utils

Change to your CUDA version

CUDA_VERSION=cu121
pip install 'vllm==0.6.1' --extra-index-url https://download.pytorch.org/whl/${CUDA_VERSION}

ntoxeg

Jan 7, 2025

Ok, after setting up an env with instructions from your last post (and vLMM 0.6.6.post1) it worked, strange that it needed a specific transformers version. Thank you for assistance.

ntoxeg changed discussion status to closed Jan 7, 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