OSWorld
Benchmark for multimodal computer-use agents on open-ended tasks in real desktop environments.
Updated
✨ Try AgentScout's free AI image generator →Good for: controlling a computer, evaluating agents
Use cases
About OSWorld
OSWorld is listed in our evals & observability category and is aimed at builders. Benchmark for multimodal computer-use agents on open-ended tasks in real desktop environments. Pricing, as the maker states it: Open source. Maturity: Active. We last checked the official link on 2026-10-18.
OSWorld at a glance
- Category: Evals & observability
- Built for: Developers and teams building with AI agents
- Pricing: Open source (as the maker states it)
- Licence: Apache License 2.0 (OSI-approved open source)
- GitHub: xlang-ai/OSWorld · 3,229 stars (as of Oct 21, 2026)
- Latest release: v0.1.16, Jun 26, 2024
- Last commit: Oct 21, 2026
- Official site: github.com/xlang-ai/OSWorld
- Listed on AgentScout: Oct 18, 2026
OSWorld FAQ
What is OSWorld?
Benchmark for multimodal computer-use agents on open-ended tasks in real desktop environments.
How much does OSWorld cost?
OSWorld's pricing, as the maker states it: Open source. Open-source tools are free to use, though you may still pay for model API usage or a hosted plan. Check the official site for current plans.
Is OSWorld open source?
Yes. OSWorld's GitHub repository is published under the Apache License 2.0, according to the licence file in the repo.
Where can I find OSWorld's source code?
On GitHub at xlang-ai/OSWorld, which had 3,229 stars as of Oct 21, 2026. The latest release is v0.1.16 (Jun 26, 2024).
What are alternatives to OSWorld?
Similar tools include SWE-bench, Galileo Agent Leaderboard, Hamming AI, Latitude, Basalt and Agenta.
Facts from our link-checked listing, the GitHub repository and the official site. See something wrong? Suggest an edit.
OSWorld ratings and reviews
Be the first to rate OSWorld: tap 1 to 5 stars below. A written review is optional.
OSWorld alternatives
Other evals & observability tools in our directory. See all evals & observability tools →
SWE-bench
Benchmark that tests whether language models and coding agents can resolve real GitHub issues, with multimodal and verified variants.
Galileo Agent Leaderboard
Galileo's leaderboard that ranks how well LLMs perform as agents on tool use and multi-step tasks.
Hamming AI
Testing and QA platform for voice and chat agents that generates scenarios and replays real calls.
Latitude
Observability platform for AI agents that captures traces, groups failures into signals and verifies fixes against real traces.
Basalt
Captures customer behavior so your AI agent can learn and improve from real interactions.
Agenta
Open-source LLMOps platform for prompt management, evaluation and observability.
Popular searches with this tool
🏆 Ranked in: Best AI Agent Observability & Eval Tools
🧭 Task guides: AI eval and benchmark tools
Badge for makers
Built OSWorld? Add this badge to your README or site. It links back to this page. About badges
