OpenCodePapers

computer-use-agents-on-mmb-gui-mmbench-gui

AgentsComputer Use Agents
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeAccuracyModelNameReleaseDate
MAI-UI Technical Report: Real-World Centric Foundation GUI Agents✓ Link91.3MAI-UI-32B2025-12-26
UI-Venus-1.5 Technical Report✓ Link88.6UI-Venus-1.5-30B-A3B2026-02-09
POINTS-GUI-G: GUI-Grounding Journey✓ Link87.0POINTS-GUI-G-8B2026-02-06
Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents✓ Link86.84GUI-Owl-1.5-32B-Instruct2026-02-15
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning✓ Link86.8GroundGUI-8B2026-10-13
Step-GUI Technical Report✓ Link85.6Step-GUI-8B2025-12-17
Mobile-Agent-v3: Fundamental Agents for GUI Automation✓ Link82.97GUI-Owl-32B2025-08-21
ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data✓ Link82.0ScaleCUA-32B2025-09-18
Grounding Computer Use Agents on Human Demonstrations✓ Link81.1GroundNext-7B (RL)2025-11-10
InfiGUI-G1: Advancing GUI Grounding with Adaptive Exploration Policy Optimization✓ Link80.8InfiGUI-G1-7B (extra data)2025-08-07
DeskForge: Dense Supervision from Desktop Environments for Computer-Use Agents✓ Link76.82DeskForge-Qwen3.5-4B2026-10-01
Check out how to contribute new results. Then edit this file on GitLab or GitHub.