ScreenSpot-Pro
ScreenSpot-Pro results reported for 1 model across 2 versions. Updated 2026-10-02.
Report an issueVersions
2 versions · Metric: Point-in-box accuracy
| Version | Released | Models | Results |
|---|---|---|---|
| ScreenSpot-Pro No toolsConfiguration: No tools | April 8, 2026 | 1 | 1 |
| ScreenSpot-Pro With PythonConfiguration: With Python | April 8, 2026 | 1 | 1 |
Recently reported results: ScreenSpot-Pro No tools
Most recent configuration: ScreenSpot-Pro No tools
| Model | Score | Source |
|---|---|---|
| Muse Spark (thinking) | 72.2% | ai.meta.com for Muse Spark (thinking) on ScreenSpot-Pro No tools (opens in a new tab)Point-in-box accuracyReported Evaluation methodologyMetaSelf-reportedRecord for Muse Spark on ScreenSpot-Pro No tools |