UI-Venus-2 Report - MemGUI: leaderboard
MemGUI-Bench tests whether mobile GUI agents retain short-term information, accumulate long-term experience and generalize across sessions and apps; main-results pass@1. From Ant Group's UI-Venus-2 technical report, keeping only runs the authors performed or reproduced; percent, higher is better.
Metric: Pass@1 (%). Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.0 Pro | 65.6 | #122 |
| 2 | Qwen 3.6 27B | 25.7 | #182 |
| 3 | Qwen 3.5 9B | 6.2 | #391 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ui-venus-2-report-memgui · How It Works · Data refreshed daily, snapshot 2026-11-05.