Dataset · xlangai
OpenCUA desktop agent trajectories
xlangai/AgentNet
22.6K human-annotated tasks for computer-use agents on Windows, macOS, and Ubuntu.
What's inside
AgentNet is the first large-scale dataset of desktop computer-use agent trajectories: 22.6K human-annotated tasks for Windows, macOS, and Ubuntu. The data is designed for training and evaluating vision-language-action (VLA) computer-use models. Language is English, modality is image-text-to-text.
What you can do with it
- Training VLA for computer use
- Evaluating desktop agents
- Fine-tuning OpenCUA
- Tasks
- Image text → text
- Languages
- English
- Downloads this month
- 14.8K
- Likes
- 99
Models trained on it
- xlangai/OpenCUA-7BImage text → text 8.7K
- xlangai/OpenCUA-32BImage text → text 279
- xlangai/OpenCUA-72BImage text → text 261
- sujitvasanth/OpenCUA7BQfp164bitImage text → text 190
- KeytechSRL/TernaryPilot-v2-Computer-Browser-Use-LoRAImage text → text 159
- sodan/laya-cua 95
- adriandj3/TernaryPilot-v1-Computer-Browser-Use-LoRAImage text → text 88
- manpk-ai/OpenCUA-7BImage text → text 32
How to load it
from datasets import load_dataset
ds = load_dataset("xlangai/AgentNet")