OSWorld 2.1 (partial)
Progress Over Time
Interactive timeline showing model performance evolution on OSWorld 2.1 (partial)
OSWorld 2.1 (partial) Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Anthropic | — | 1.0M | $2.00 / $10.00 | ||
| 2 |  Meta | — | — | — | ||
| 2 | Meta | — | — | — | ||
| 4 | Meta | — | — | — | ||
| 4 | Meta | — | 1.0M | $4.00 / $20.00 | ||
| 6 | Anthropic | — | 1.0M | $0.10 / $0.50 |
What is OSWorld 2.1 (partial)?
Partial-credit computer-use score on OSWorld 2.1, as reported in the Claude Sonnet 5.5 release. Separate from OSWorld 2.0 and strict completion scores.
OSWorld 2.1 (partial) is a multimodal benchmark evaluating models on multimodal, general, agents, and vision tasks. LLM Stats tracks 6 models on this benchmark, scored on a 0–1 scale. The current average is 0.8, with the leader at 0.8.
Compare leaders on the best AI for multimodal, best AI for general, best AI for agents and best AI for vision leaderboards.
Current leaders
Claude Sonnet 5.5 from Anthropic currently leads the OSWorld 2.1 (partial) leaderboard with a score of 0.801 across 6 evaluated AI models.
FAQ
Common questions about the OSWorld 2.1 (partial) benchmark and leaderboard.