{"version":1,"type":"story","url":"https://digestai.news/story/prentisai-releases-groundgui-8b-model-for-gui-agent-tasks","json":"https://digestai.news/story/prentisai-releases-groundgui-8b-model-for-gui-agent-tasks.json","markdown":"https://digestai.news/story/prentisai-releases-groundgui-8b-model-for-gui-agent-tasks.md","slug":"prentisai-releases-groundgui-8b-model-for-gui-agent-tasks","headline":"PrentisAI releases GroundGUI-8B model for GUI agent tasks","summary":"PrentisAI introduced **GroundGUI-8B**, a multimodal model designed for precise GUI agent interactions. The model uses **ScreenRef**, a dataset of 586,000 screenshots and 633,000 referring expressions across desktop, web, and mobile environments. It combines supervised fine-tuning with **Distillation-to-Reinforcement Learning (D2RL)** to improve generalization and address sparse rewards in reinforcement learning.\n\nGroundGUI-8B outperforms 8B-scale models on benchmarks like **ScreenSpot-V2 (95.2%)** and **MMBench-GUI L2 (86.8%)**, matching or exceeding performance of models trained on proprietary data. The project includes open-source code, models, and datasets, available at [GitHub](https://github.com/PrentisAI/GroundGUI).","keyPoints":["GroundGUI-8B model achieves 95.2% accuracy on ScreenSpot-V2 benchmark, surpassing 8B-scale competitors","ScreenRef dataset contains 586K screenshots and 633K referring expressions for GUI grounding tasks","D2RL method combines supervised fine-tuning with reinforcement learning for improved generalization"],"whyItMatters":"GroundGUI-8B advances GUI agent capabilities, enabling more accurate natural language-to-action mapping in software interfaces. This could accelerate automation and accessibility for developers and users alike.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["PrentisAI"],"models":["GroundGUI-8B","Qwen3-VL-8B-Instruct"],"people":[]},"firstPublishedAt":"2026-10-15T04:00:00Z","updatedAt":"2026-10-15T04:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"arXiv cs.AI","title":"GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning","url":"https://arxiv.org/abs/2610.19100","publishedAt":"2026-10-15T04:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"PrentisAI releases GroundGUI-8B model for GUI agent tasks\", 15 October 2026, https://digestai.news/story/prentisai-releases-groundgui-8b-model-for-gui-agent-tasks","publisher":"Digest AI","title":"PrentisAI releases GroundGUI-8B model for GUI agent tasks","datePublished":"2026-10-15T04:00:00Z","url":"https://digestai.news/story/prentisai-releases-groundgui-8b-model-for-gui-agent-tasks"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}