OpenGCUA OSWorld-v2 offline GUI batch success: leaderboard
OpenGCUA paper, OSWorld-v2 v2.1 82-task offline subset excluding26 internet tasks. Latest evaluated attempt per task, never best-of selection; failures retained. Source model identities as printed; effort unspecified for aggregate tables. Partial evaluator score and full-task success are separate metrics. GUI batch GUI-only batched actions. Metric: Success (%). Higher is better. Source: https://arxiv.org/html/2610.22200v1
Metric: Success (%). Source: arxiv.org. Saturation forecast: Around March 2028. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 36.6 | #5 |
| 2 | Claude Opus 5 | 29.3 | #12 |
| 3 | Gemini 3.8 Flash | 22 | #21 |
| 4 | Qwen 3.8 Max | 17.1 | #31 |
| 5 | Kimi K3 | 15.9 | #29 |
| 6 | GLM-5.3 Flash | 14.6 | #62 |
| 7 | Qwen 3.8 Flash | 14.2 | #47 |
| 8 | DeepSeek V4.1 Flash | 8.5 | #35 |
No result here: #3 Claude Opus 5.5, #7 Gemini 4 Argon, #8 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opengcua-osworld-v2-offline-gui-batch-success · How It Works · Data refreshed daily, snapshot 2026-10-17.