GroundGUI-8B

A GUI grounding model built on Qwen3-VL-8B-Instruct. Given a screenshot and a natural-language instruction, it outputs a single click action as a <tool_call> with a coordinate on a 0–1000 grid.

• Code: PrentisAI/GroundGUI
• Data: PrentisAI/ScreenRef

Training

Fine-tuned from Qwen3-VL-8B-Instruct on ScreenRef. Details will be described in the paper.

Results

Accuracy (%) is the fraction of predicted click points that land inside the ground-truth box. Coordinates are mapped back to pixel space and checked against the box exactly, with no rounding tolerance.

• Inference used vLLM with greedy decoding (temperature 0) and max_tokens 1024.
• OSWorld-G is the original 564-question split.

Usage

The model expects a specific system prompt, and the chat template does not add it for you. Use the evaluation code in GroundGUI (scripts/eval.sh) to reproduce the numbers above; it sets the system prompt (scalecua_toolcall) and the scoring for each benchmark.

The output looks like:

<tool_call>
{"name": "computer_use", "arguments": {"action": "left_click", "coordinate": [x, y]}}
</tool_call>

To get pixel coordinates from the output: x_px = x / 1000 * image_width and y_px = y / 1000 * image_height.

Notes

• The tool name depends on the platform: computer_use for desktop, browser_use for web, mobile_use for mobile.
• generation_config.json is inherited from the base model and enables sampling. Pass do_sample=False to reproduce the numbers above.
• The weights were saved with transformers 5.12.

License

Apache License 2.0, the same as the base model. This model is a modified version of Qwen3-VL-8B-Instruct: Prentis AI fine-tuned the original weights.

The training data, ScreenRef, is distributed under its own terms; this license covers the model only and grants no rights in the dataset or in the screenshots it contains.

Citation

If you use this model, please cite the ScreenRef dataset: PrentisAI/ScreenRef. A BibTeX entry will be added when the paper is public.