OSWorld
OSWorld results reported for 34 models across 8 versions. Updated 2026-10-07.
Report an issueVersions
8 versions, plus 3 variants and configurations
| Version | Released | Metric | Models | Results | ||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OSWorld 2.0 v2026.08.08 Offline — Partial, max of 3 runsLatest | September 30, 2026 | OSWorld Partial Score | 1 | 1 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.08.08 Offline — Partial | August 8, 2026 | OSWorld Partial Score | 6 | 31 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.08.08 — Binary | August 8, 2026 | OSWorld Strict Score | 1 | 1 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.08.08 — Partial | August 8, 2026 | OSWorld Partial Score | 2 | 2 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.08.08 — Strict | August 8, 2026 | OSWorld Strict Score | 1 | 1 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.06.24 — Binary | June 26, 2026 | OSWorld Strict Score | 1 | 1 | ||||||||||||||||||||||||
| OSWorld 2.0 v2026.06.24 — Partial | June 26, 2026 | OSWorld Partial Score | 1 | 1 | ||||||||||||||||||||||||
OSWorld Verified3 variants of Verified
| July 28, 2025 | OSWorld-Verified success rate | 18 | 18 |
Recently reported results: OSWorld 2.0 v2026.08.08 Offline — Partial, max of 3 runs
Latest version: OSWorld 2.0 v2026.08.08 Offline — Partial, max of 3 runs
| Model | Score | Source |
|---|---|---|
| Gemini 4 Argon (highest thinking settings) | 69.2% | deepmind.google for Gemini 4 Argon (highest thinking settings) on OSWorld 2.0 v2026.08.08 Offline — Partial, max of 3 runs (opens in a new tab)OSWorld Partial ScoreReported Evaluation methodologyGoogle DeepMindSelf-reportedRecord for Gemini 4 Argon on OSWorld 2.0 v2026.08.08 Offline — Partial, max of 3 runs |