Dataset · webarena-x
WebArena web agent trajectories
webarena-x/webarena-infinity-trajectories
Dataset of successful browser agent trajectories with step-by-step screenshots and reasoning for agent training.
What's inside
Contains successful trajectories for executing real tasks on web applications from the WebArena-Infinity benchmark. Size ranges from 10K to 100K entries. Each entry includes step-by-step screenshots, agent reasoning, and performed actions.
What you can do with it
- Training and evaluating universal web agents
- Developing models for visual analysis of web interfaces
- Tasks
- Visual question answering, Image → text
- Size
- 10K–100K rows
- Downloads this month
- 20.2K
- Likes
- 2
How to load it
from datasets import load_dataset
ds = load_dataset("webarena-x/webarena-infinity-trajectories")