OSWorld 2.1 D
AI benchmark · Agentic
What is OSWorld 2.1?
OSWorld 2.1 measures how well AI agents complete long-horizon computer-use tasks across desktop apps, mocked websites and GitLab in real VM environments.
Computer use mostly
“OSWorld 2.1 can run with different environment providers depending on where you test”
github.com/xlang-ai/OSWorld-V2 ↗
Who built OSWorld 2.1
Led by XLang Lab (xlang-ai).
Lineage
Benchmarks this one is built on and built from: versions, subsets, extensions and the indexes that combine them.
Built on
OSWorld earlier version
OSWorld 2.0 earlier version
Tier D
Long tail: little or no usage data yet.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- no usage data
- in curated RL directories (1 pt)
On X · well regarded, moderate buzz, from 9 posts about OSWorld 2.1 in the last month
OSWorld 2.1 is the headline computer-use metric in Anthropic's Haiku 5.5 and OpenAI's GPT-6.1 Terra launches, widely repeated without criticism of the benchmark.
How we got to tier D
- Adoption: no usage data. Tier D.
- On X: well regarded, moderate buzz. “OSWorld 2.1 is the headline computer-use metric in Anthropic's Haiku 5.5 and OpenAI's GPT-6.1 Terra launches, widely repeated without criticism of the benchmark.” No move.
- In curated RL directories: 1 point, a floor of none.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where OSWorld 2.1 sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-29 snapshot.