pip install transformers
pip install qwen-vl-utils

from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
from qwen_vl_utils import process_vision_info

# Default: Load the model on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained(
    "HongxinLi/UIPro-7B_Stage2_Mobile", torch_dtype="auto", device_map="auto"
)
processor = AutoProcessor.from_pretrained("HongxinLi/UIPro-7B_Stage2_Mobile")

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "./web_6f93090a-81f6-489e-bb35-1a2838b18c01.png",
            },

            {"type": "text", "text": """Given the Web UI screenshot and previous actions, please generate the next move necessary to advance towards task completion. The user's task is: {task}
Action history: {action_history}

Now, first describe the action intent and then directly plan the next action."""},
        ],
    }
]

@InProceedings{Li_2025_ICCV,
    author    = {Li, Hongxin and Su, Jingran and Chen, Jingfan and Ju, Zheng and Chen, Yuntao and Li, Qing and Zhang, Zhaoxiang},
    title     = {UIPro: Unleashing Superior Interaction Capability For GUI Agents},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2025},
    pages     = {1613-1623}
}