# OSWorld 2.1 (full set) — AI model rankings

An agent operates a desktop across 108 tasks. Each completed checkpoint earns part of the task’s credit, so a model can score even when it only finishes part of a task.

5 tracked models have a published OSWorld 2.1 (full set) score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Claude Opus 5.5 | Anthropic | 81.8% | Lab | Sep 22 2026 |
| 2 | Claude Fable 5.1 | Anthropic | 80.7% | Lab | Sep 1 2026 |
| 3 | Claude Sonnet 5.5 | Anthropic | 80.1% | Lab | Sep 28 2026 |
| 4 | Claude Opus 5 | Anthropic | 74% | Lab | Jul 24 2026 |
| 5 | Claude Sonnet 5 | Anthropic | 57% | Lab | Jun 30 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/osworld-2.1
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.