← all repositories

microsoft/WindowsAgentArena

Windows Agent Arena is a Microsoft platform for testing and benchmarking multi-modal desktop AI agents on the Windows OS.

★912 stars Python LLMOps · EvalAgents
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history

Windows Agent Arena provides a reproducible Windows OS environment where agentic AI workflows can be evaluated across a diverse range of desktop tasks. It supports deploying agents at scale on Azure ML infrastructure, enabling parallel benchmark runs that produce results for hundreds of tasks in minutes. The project is aimed at researchers and developers building and evaluating multi-modal, computer-use agents.

Frequently asked

What is microsoft/WindowsAgentArena?
Windows Agent Arena is a Microsoft platform for testing and benchmarking multi-modal desktop AI agents on the Windows OS.
Is WindowsAgentArena open source?
Yes — microsoft/WindowsAgentArena is open source, released under the MIT license.
What language is WindowsAgentArena written in?
microsoft/WindowsAgentArena is primarily written in Python.
How popular is WindowsAgentArena?
microsoft/WindowsAgentArena has 912 stars on GitHub.
Where can I find WindowsAgentArena?
microsoft/WindowsAgentArena is on GitHub at https://github.com/microsoft/WindowsAgentArena.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.