web-arena-x/visualwebarena

VisualWebArena is a benchmark for multimodal agents.

47
/ 100
Emerging

This is a benchmark for anyone building or researching multimodal AI agents that interact with websites. It helps you test how well your agent can perform complex, realistic web tasks by providing a standardized set of challenges and evaluation scripts. You put in your multimodal agent and get out detailed performance metrics on how accurately it completes various tasks like online shopping, social media interactions, or information retrieval.

445 stars. No commits in the last 6 months.

Use this if you are developing or evaluating autonomous AI agents that need to understand and interact with visual elements on websites to complete multi-step tasks.

Not ideal if you are looking for an off-the-shelf tool for general web automation or if your AI agent doesn't require visual comprehension of web pages.

AI agent development multimodal AI evaluation web automation research autonomous systems testing natural language processing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

445

Forks

72

Language

Python

License

MIT

Last pushed

Nov 09, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/web-arena-x/visualwebarena"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.