|
|
--- |
|
|
base_model: |
|
|
- Qwen/Qwen2.5-VL-7B-Instruct |
|
|
language: |
|
|
- en |
|
|
library_name: transformers |
|
|
license: apache-2.0 |
|
|
pipeline_tag: image-text-to-text |
|
|
tags: |
|
|
- gui |
|
|
- agent |
|
|
- gui-grounding |
|
|
- reinforcement-learning |
|
|
--- |
|
|
|
|
|
# InfiGUI-G1-7B |
|
|
|
|
|
This repository contains the InfiGUI-G1-7B model from the paper **[InfiGUI-G1: Advancing GUI Grounding with Adaptive Exploration Policy Optimization](https://arxiv.org/abs/2508.05731)**. |
|
|
|
|
|
<p align="left"> |
|
|
<a href="https://arxiv.org/abs/2508.05731"><img src="https://img.shields.io/badge/arXiv-Preprint-b31b1b?style=flat&logo=arxiv&logoColor=white" alt="arXiv Paper"></a> |
|
|
<a href="https://huggingface.co/papers/2508.05731"><img src="https://img.shields.io/badge/HuggingFace-Daily%20Papers-ff9800?style=flat&logo=huggingface" alt="Hugging Face Paper"></a> |
|
|
<a href="https://huggingface.co/InfiX-ai/InfiGUI-G1-3B"><img src="https://img.shields.io/badge/Model-InfiGUI--G1--3B-007ec6?style=flat&logo=huggingface" alt="InfiGUI-G1 3B Model"></a> |
|
|
<a href="https://github.com/InfiXAI/InfiGUI-G1"><img src="https://img.shields.io/badge/GitHub-Repo-181717?style=flat&logo=github&logoColor=white" alt="GitHub Repo"></a> |
|
|
</p> |
|
|
|
|
|
## Model Description |
|
|
|
|
|
The model is based on `Qwen2.5-VL-7B-Instruct` and is fine-tuned using our proposed **Adaptive Exploration Policy Optimization (AEPO)** framework. AEPO is a novel reinforcement learning method designed to enhance the model's **semantic alignment** for GUI grounding tasks. It overcomes the exploration bottlenecks of standard RLVR methods by integrating a multi-answer generation strategy with a theoretically-grounded adaptive reward function, enabling more effective and efficient learning for complex GUI interactions. |
|
|
|
|
|
## Paper Overview |
|
|
|
|
|
A fundamental challenge for GUI agents is robustly grounding natural language instructions, which requires not only precise **spatial alignment** (locating elements accurately) but also correct **semantic alignment** (identifying the functionally appropriate element). While existing Reinforcement Learning with Verifiable Rewards (RLVR) methods have enhanced spatial precision, they often suffer from inefficient exploration. This "confidence trap" bottlenecks semantic alignment, preventing models from discovering correct actions for difficult semantic associations. |
|
|
|
|
|
To address this critical exploration problem, we introduce **InfiGUI-G1**, a series of models trained with **Adaptive Exploration Policy Optimization (AEPO)**. AEPO overcomes the exploration bottleneck by integrating a **multi-answer generation** strategy to explore a diverse set of candidate actions in a single forward pass. This exploration is guided by a theoretically-grounded **Adaptive Exploration Reward (AER)** function, derived from first principles of efficiency (η=U/C), which provides rich, informative learning signals to dynamically balance exploration and exploitation. |
|
|
|
|
|
## Quick Start |
|
|
|
|
|
### Installation |
|
|
First, install the required dependencies: |
|
|
```bash |
|
|
pip install transformers qwen-vl-utils |
|
|
```` |
|
|
|
|
|
### Example |
|
|
|
|
|
```python |
|
|
import json |
|
|
import math |
|
|
import torch |
|
|
import requests |
|
|
from io import BytesIO |
|
|
from PIL import Image, ImageDraw, ImageFont |
|
|
from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessor |
|
|
from qwen_vl_utils import process_vision_info, smart_resize |
|
|
|
|
|
MAX_IMAGE_PIXELS = 5600 * 28 * 28 |
|
|
|
|
|
|
|
|
def resize_image(width: int, height: int, max_pixels: int) -> tuple[int, int]: |
|
|
""" |
|
|
Resize image to fit within max_pixels constraint while maintaining aspect ratio. |
|
|
Applies smart_resize for final dimension optimization. |
|
|
""" |
|
|
current_pixels = width * height |
|
|
|
|
|
if current_pixels <= max_pixels: |
|
|
target_width, target_height = width, height |
|
|
else: |
|
|
scale_factor = math.sqrt(max_pixels / current_pixels) |
|
|
target_width = round(width * scale_factor) |
|
|
target_height = round(height * scale_factor) |
|
|
|
|
|
# Apply smart_resize for final dimensions |
|
|
final_height, final_width = smart_resize(target_height, target_width) |
|
|
|
|
|
return final_width, final_height |
|
|
|
|
|
|
|
|
def load_image(img_path: str) -> Image.Image: |
|
|
"""Load image from URL or local path.""" |
|
|
if img_path.startswith("https://"): |
|
|
response = requests.get(img_path) |
|
|
return Image.open(BytesIO(response.content)) |
|
|
else: |
|
|
return Image.open(img_path) |
|
|
|
|
|
|
|
|
def visualize_points(original_image: Image.Image, points: list, |
|
|
new_width: int, new_height: int, |
|
|
original_width: int, original_height: int) -> None: |
|
|
"""Draw prediction points on original image and save as output.png.""" |
|
|
output_img = original_image.copy() |
|
|
draw = ImageDraw.Draw(output_img) |
|
|
font = ImageFont.load_default(size=100) |
|
|
|
|
|
for i, point_data in enumerate(points): |
|
|
coords = point_data['point_2d'] |
|
|
|
|
|
# Map coordinates from resized image back to original image |
|
|
original_x = int(coords[0] / new_width * original_width) |
|
|
original_y = int(coords[1] / new_height * original_height) |
|
|
|
|
|
label = str(i + 1) |
|
|
|
|
|
# Draw circle |
|
|
circle_radius = 20 |
|
|
draw.ellipse([original_x - circle_radius, original_y - circle_radius, |
|
|
original_x + circle_radius, original_y + circle_radius], |
|
|
fill=(255, 0, 0)) |
|
|
|
|
|
# Draw label |
|
|
draw.text((original_x + 20, original_y - 20), label, fill=(255, 0, 0), font=font) |
|
|
|
|
|
print(f"Point {i+1}: Predicted coordinates {coords} -> Mapped coordinates [{original_x}, {original_y}]") |
|
|
|
|
|
output_img.save("output.png") |
|
|
print(f"Visualization with {len(points)} points saved to output.png") |
|
|
|
|
|
|
|
|
def main(): |
|
|
# Load model and processor |
|
|
model = Qwen2_5_VLForConditionalGeneration.from_pretrained( |
|
|
"InfiX-ai/InfiGUI-G1-7B", |
|
|
torch_dtype=torch.bfloat16, |
|
|
attn_implementation="flash_attention_2", |
|
|
device_map="auto" |
|
|
) |
|
|
processor = AutoProcessor.from_pretrained("InfiX-ai/InfiGUI-G1-7B", padding_side="left") |
|
|
|
|
|
# Load and process image |
|
|
img_path = "https://raw.githubusercontent.com/InfiXAI/InfiGUI-G1/main/assets/test_image.png" |
|
|
image = load_image(img_path) |
|
|
|
|
|
# Store original image and resize for model input |
|
|
original_image = image.copy() |
|
|
original_width, original_height = image.size |
|
|
new_width, new_height = resize_image(original_width, original_height, MAX_IMAGE_PIXELS) |
|
|
resized_image = image.resize((new_width, new_height)) |
|
|
|
|
|
# Prepare model inputs |
|
|
instruction = "shuffle play the current playlist" |
|
|
system_prompt = 'You FIRST think about the reasoning process as an internal monologue and then provide the final answer.\nThe reasoning process MUST BE enclosed within <think> </think> tags.' |
|
|
prompt = f'''The screen's resolution is {new_width}x{new_height}. |
|
|
Locate the UI element(s) for "{instruction}", output the coordinates using JSON format: [{{"point_2d": [x, y]}}, ...]''' |
|
|
|
|
|
messages = [ |
|
|
{"role": "system", "content": system_prompt}, |
|
|
{ |
|
|
"role": "user", |
|
|
"content": [ |
|
|
{"type": "image", "image": resized_image}, |
|
|
{"type": "text", "text": prompt} |
|
|
] |
|
|
} |
|
|
] |
|
|
|
|
|
# Generate predictions |
|
|
text = processor.apply_chat_template([messages], tokenize=False, add_generation_prompt=True) |
|
|
image_inputs, video_inputs = process_vision_info([messages]) |
|
|
inputs = processor(text=text, images=image_inputs, videos=video_inputs, padding=True, return_tensors="pt").to("cuda") |
|
|
generated_ids = model.generate(**inputs, max_new_tokens=512) |
|
|
output_text = processor.batch_decode( |
|
|
[out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)], |
|
|
skip_special_tokens=True, |
|
|
clean_up_tokenization_spaces=False |
|
|
) |
|
|
|
|
|
# Parse and visualize results |
|
|
output_text = output_text[0].split("</think>")[-1].replace("```json", "").replace("```", "").strip() |
|
|
output = json.loads(output_text) |
|
|
|
|
|
if output: |
|
|
visualize_points(original_image, output, new_width, new_height, original_width, original_height) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
main() |
|
|
``` |
|
|
|
|
|
## Results |
|
|
|
|
|
Our InfiGUI-G1 models, trained with the AEPO framework, establish new state-of-the-art results among open-source models across a diverse and challenging set of GUI grounding benchmarks: |
|
|
|
|
|
<div align="left"> |
|
|
<table style="width: 100%; max-width: 750px; border-collapse: collapse; border-top: 2px solid #212529; border-bottom: 2px solid #212529; font-family: sans-serif;"> |
|
|
<thead style="background-color: #f8f9fa;"> |
|
|
<tr style="border-bottom: 1.5px solid #212529;"> |
|
|
<th style="padding: 12px 10px; text-align: left; width: 24.9%; font-weight: 600; color: #343a40;">Model</th> |
|
|
<th style="padding: 12px 10px; text-align: center; font-weight: 600; color: #343a40;">MMBench-GUI</th> |
|
|
<th style="padding: 12px 10px; text-align: center; font-weight: 600; color: #343a40;">ScreenSpot-v2</th> |
|
|
<th style="padding: 12px 10px; text-align: center; font-weight: 600; color: #343a40;">UI-Vision</th> |
|
|
<th style="padding: 12px 10px; text-align: center; font-weight: 600; color: #343a40;">I2E-Bench</th> |
|
|
<th style="padding: 12px 10px; text-align: center; font-weight: 600; color: #343a40;">ScreenSpot-Pro</th> |
|
|
</tr> |
|
|
</thead> |
|
|
<tbody> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">Qwen2.5-VL-7B</td> |
|
|
<td style="padding: 10px; text-align: center;">33.9</td> |
|
|
<td style="padding: 10px; text-align: center;">88.8</td> |
|
|
<td style="padding: 10px; text-align: center;">0.9</td> |
|
|
<td style="padding: 10px; text-align: center;">53.8</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">GUI-G²-7B</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;"><u>93.3</u></td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">47.5</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">UI-TARS-7B</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">91.6</td> |
|
|
<td style="padding: 10px; text-align: center;">17.6</td> |
|
|
<td style="padding: 10px; text-align: center;">61.4</td> |
|
|
<td style="padding: 10px; text-align: center;">35.7</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">UGround-v1-7B</td> |
|
|
<td style="padding: 10px; text-align: center;">65.7</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">12.9</td> |
|
|
<td style="padding: 10px; text-align: center;">70.3</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">UI-TARS-1.5-7B</td> |
|
|
<td style="padding: 10px; text-align: center;">64.3</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">73.2</td> |
|
|
<td style="padding: 10px; text-align: center;"><u>49.6</u></td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">Qwen2.5-VL-72B</td> |
|
|
<td style="padding: 10px; text-align: center;">41.8</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">51.4</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">UGround-v1-72B</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
<td style="padding: 10px; text-align: center;">23.2</td> |
|
|
<td style="padding: 10px; text-align: center;"><u>76.3</u></td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<td style="padding: 10px; text-align: left;">UI-TARS-72B</td> |
|
|
<td style="padding: 10px; text-align: center;"><u>74.3</u></td> |
|
|
<td style="padding: 10px; text-align: center;">90.3</td> |
|
|
<td style="padding: 10px; text-align: center;"><u>25.5</u></td> |
|
|
<td style="padding: 10px; text-align: center;">73.7</td> |
|
|
<td style="padding: 10px; text-align: center;">-</td> |
|
|
</tr> |
|
|
<tr> |
|
|
<th colspan="6" style="padding: 10px 12px; text-align: left; font-style: italic; background-color: #f8f9fa; border-top: 1px solid #dee2e6; border-bottom: 1px solid #dee2e6; color: #343a40;">Ours</th> |
|
|
</tr> |
|
|
<tr style="background-color: #f0f8ff;"> |
|
|
<td style="padding: 10px; text-align: left;"><b>InfiGUI-G1-7B</b></td> |
|
|
<td style="padding: 10px; text-align: center;"><b>80.8</b></td> |
|
|
<td style="padding: 10px; text-align: center;"><b>93.5</b></td> |
|
|
<td style="padding: 10px; text-align: center;"><b>26.1</b></td> |
|
|
<td style="padding: 10px; text-align: center;"><b>77.4</b></td> |
|
|
<td style="padding: 10px; text-align: center;"><b>51.9</b></td> |
|
|
</tr> |
|
|
<tr style="background-color: #f0f8ff;"> |
|
|
<td style="padding: 10px; text-align: right;"><i>w/ Expl. Success</i></td> |
|
|
<td style="padding: 10px; text-align: center;">86.4</td> |
|
|
<td style="padding: 10px; text-align: center;">95.6</td> |
|
|
<td style="padding: 10px; text-align: center;">34.4</td> |
|
|
<td style="padding: 10px; text-align: center;">83.0</td> |
|
|
<td style="padding: 10px; text-align: center;">58.0</td> |
|
|
</tr> |
|
|
</tbody> |
|
|
</table> |
|
|
</div> |
|
|
|
|
|
## Evaluation |
|
|
|
|
|
This section provides instructions for reproducing the evaluation results reported in our paper. |
|
|
|
|
|
### 1. Getting Started |
|
|
|
|
|
Clone the repository and navigate to the project directory: |
|
|
|
|
|
```bash |
|
|
git clone https://github.com/InfiXAI/InfiGUI-G1.git |
|
|
cd InfiGUI-G1 |
|
|
``` |
|
|
|
|
|
### 2. Environment Setup |
|
|
|
|
|
The evaluation pipeline is built upon the [vLLM](https://github.com/vllm-project/vllm) library for efficient inference. For detailed installation guidance, please refer to the official vLLM repository. The specific versions used to obtain the results reported in our paper are as follows: |
|
|
|
|
|
- **Python**: `3.10.12` |
|
|
- **PyTorch**: `2.6.0` |
|
|
- **Transformers**: `4.50.1` |
|
|
- **vLLM**: `0.8.2` |
|
|
- **CUDA**: `12.6` |
|
|
|
|
|
The reported results were obtained on a server equipped with 4 x NVIDIA H800 GPUs. |
|
|
|
|
|
### 3. Model Download |
|
|
|
|
|
Download the InfiGUI-G1 models from the Hugging Face Hub into the `./models` directory. |
|
|
|
|
|
```bash |
|
|
# Create a directory for models |
|
|
mkdir -p ./models |
|
|
|
|
|
# Download InfiGUI-G1-3B |
|
|
huggingface-cli download --resume-download InfiX-ai/InfiGUI-G1-3B --local-dir ./models/InfiGUI-G1-3B |
|
|
|
|
|
# Download InfiGUI-G1-7B |
|
|
huggingface-cli download --resume-download InfiX-ai/InfiGUI-G1-7B --local-dir ./models/InfiGUI-G1-7B |
|
|
``` |
|
|
|
|
|
### 4. Dataset Download and Preparation |
|
|
|
|
|
Download the required evaluation benchmarks into the `./data` directory. |
|
|
|
|
|
```bash |
|
|
# Create a directory for datasets |
|
|
mkdir -p ./data |
|
|
|
|
|
# Download benchmarks |
|
|
huggingface-cli download --repo-type dataset --resume-download likaixin/ScreenSpot-Pro --local-dir ./data/ScreenSpot-Pro |
|
|
huggingface-cli download --repo-type dataset --resume-download ServiceNow/ui-vision --local-dir ./data/ui-vision |
|
|
huggingface-cli download --repo-type dataset --resume-download OS-Copilot/ScreenSpot-v2 --local-dir ./data/ScreenSpot-v2 |
|
|
huggingface-cli download --repo-type dataset --resume-download OpenGVLab/MMBench-GUI --local-dir ./data/MMBench-GUI |
|
|
huggingface-cli download --repo-type dataset --resume-download vaundys/I2E-Bench --local-dir ./data/I2E-Bench |
|
|
``` |
|
|
|
|
|
After downloading, some datasets require unzipping compressed image files. |
|
|
|
|
|
```bash |
|
|
# Unzip images for ScreenSpot-v2 |
|
|
unzip ./data/ScreenSpot-v2/screenspotv2_image.zip -d ./data/ScreenSpot-v2/ |
|
|
|
|
|
# Unzip images for MMBench-GUI |
|
|
unzip ./data/MMBench-GUI/MMBench-GUI-OfflineImages.zip -d ./data/MMBench-GUI/ |
|
|
``` |
|
|
|
|
|
### 5. Running the Evaluation |
|
|
|
|
|
To run the evaluation, use the `eval/eval.py` script. You must specify the path to the model, the benchmark name, and the tensor parallel size. |
|
|
|
|
|
Here is an example command to evaluate the `InfiGUI-G1-3B` model on the `screenspot-pro` benchmark using 4 GPUs: |
|
|
|
|
|
```bash |
|
|
python eval/eval.py \ |
|
|
./models/InfiGUI-G1-3B \ |
|
|
--benchmark screenspot-pro \ |
|
|
--tensor-parallel 4 |
|
|
``` |
|
|
|
|
|
- **`model_path`**: The first positional argument specifies the path to the downloaded model directory (e.g., `./models/InfiGUI-G1-3B`). |
|
|
- **`--benchmark`**: Specifies the benchmark to evaluate. Available options include `screenspot-pro`, `screenspot-v2`, `ui-vision`, `mmbench-gui`, and `i2e-bench`. |
|
|
- **`--tensor-parallel`**: Sets the tensor parallelism size, which should typically match the number of available GPUs. |
|
|
|
|
|
Evaluation results, including detailed logs and performance metrics, will be saved to the `./output/{model_name}/{benchmark}/` directory. |
|
|
|
|
|
## Citation Information |
|
|
|
|
|
If you find this work useful, we would be grateful if you consider citing the following papers: |
|
|
|
|
|
```bibtex |
|
|
@misc{liu2025infiguig1advancingguigrounding, |
|
|
title={InfiGUI-G1: Advancing GUI Grounding with Adaptive Exploration Policy Optimization}, |
|
|
author={Yuhang Liu and Zeyu Liu and Shuanghe Zhu and Pengxiang Li and Congkai Xie and Jiasheng Wang and Xueyu Hu and Xiaotian Han and Jianbo Yuan and Xinyao Wang and Shengyu Zhang and Hongxia Yang and Fei Wu}, |
|
|
year={2025}, |
|
|
eprint={2508.05731}, |
|
|
archivePrefix={arXiv}, |
|
|
primaryClass={cs.AI}, |
|
|
url={https://arxiv.org/abs/2508.05731}, |
|
|
} |
|
|
``` |
|
|
|
|
|
```bibtex |
|
|
@article{liu2025infigui, |
|
|
title={InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners}, |
|
|
author={Liu, Yuhang and Li, Pengxiang and Xie, Congkai and Hu, Xavier and Han, Xiaotian and Zhang, Shengyu and Yang, Hongxia and Wu, Fei}, |
|
|
journal={arXiv preprint arXiv:2504.14239}, |
|
|
year={2025} |
|
|
} |
|
|
``` |
|
|
|
|
|
```bibtex |
|
|
@article{liu2025infiguiagent, |
|
|
title={InfiGUIAgent: A Multimodal Generalist GUI Agent with Native Reasoning and Reflection}, |
|
|
author={Liu, Yuhang and Li, Pengxiang and Wei, Zishu and Xie, Congkai and Hu, Xueyu and Xu, Xinchen and Zhang, Shengyu and Han, Xiaotian and Yang, Hongxia and Wu, Fei}, |
|
|
journal={arXiv preprint arXiv:2501.04575}, |
|
|
year={2025} |
|
|
} |
|
|
``` |
|
|
|
|
|
## Acknowledgements |
|
|
|
|
|
We would like to express our gratitude for the following open-source projects: [VERL](https://github.com/volcengine/verl), [Qwen2.5-VL](https://github.com/QwenLM/Qwen2.5-VL) and [vLLM](https://github.com/vllm-project/vllm). |
