ML Research Wiki / Benchmarks / Natural Language Visual Grounding / ScreenSpot

ScreenSpot

Natural Language Visual Grounding Benchmark

Performance Over Time

📊 Showing 18 results | 📏 Metric: Accuracy (%)

Top Performing Models

Rank Model Paper Accuracy (%) Date Code
1 UGround-V1-7B Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents 86.34 2024-10-07 📦 OSU-NLP-Group/UGround
2 Aguvis-7B Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction 83.00 2024-12-05 📦 xlang-ai/aguvis
3 OS-Atlas-Base-7B OS-ATLAS: A Foundation Action Model for Generalist GUI Agents 82.47 2024-10-30 📦 njucckevin/seeclick 📦 OS-Copilot/OS-Atlas
4 Aria-UI Aria-UI: Visual Grounding for GUI Instructions 81.10 2024-12-20 📦 ariaui/aria-ui
5 Aguvis-G-7B Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction 81.00 2024-12-05 📦 xlang-ai/aguvis
6 UGround-V1-2B Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents 77.67 2024-10-07 📦 OSU-NLP-Group/UGround
7 ShowUI ShowUI: One Vision-Language-Action Model for GUI Visual Agent 75.10 2024-11-26 📦 showlab/showui
8 ShowUI-G ShowUI: One Vision-Language-Action Model for GUI Visual Agent 75.00 2024-11-26 📦 showlab/showui
9 UGround Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents 73.30 2024-10-07 📦 OSU-NLP-Group/UGround
10 OmniParser OmniParser for Pure Vision Based GUI Agent 73.00 2024-08-01 📦 microsoft/omniparser

All Papers (18)

OSU-NLP-Group/UGround
xlang-ai/aguvis
njucckevin/seeclick OS-Copilot/OS-Atlas
ariaui/aria-ui
xlang-ai/aguvis
OSU-NLP-Group/UGround
showlab/showui
showlab/showui
OSU-NLP-Group/UGround
microsoft/omniparser
njucckevin/seeclick OS-Copilot/OS-Atlas
njucckevin/seeclick
thudm/cogvlm THUDM/CogAgent digirl-agent/digirl
qwenlm/qwen2-vl qwenlm/qwen2.5-vl
yiye3/guicourse
vision-cair/minigpt-4 zebangcheng/emotion-llama
FoundationVision/Groma
qwenlm/qwen-vl brandon3964/multimodal-task-vector