MobiFlow - App Interference Completion: leaderboard

MobiFlow (2026) tests mobile GUI agents on 240 real-world tasks in 20 third-party Android apps, scoring progress on task graphs fused from real user trajectories. Completion rate under app interference such as pop-ups, for general multimodal models in the shared framework of the paper, which marks actionable screen elements with OmniParser, in percent; higher is better.

Metric: Completion rate (%) under app interference (pop-ups and other interruptions) on MobiFlow's 240 real-world tasks in 20 third-party Android apps, scored on task-completion state graphs fused from real trajectories, at most 50 steps; general models in the paper's shared execution framework (OmniParser marks actionable elements; OpenRouter, temperature 0); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.170.1#278
2Gemini 3 Flash58.4#175
3Gemini 2.5 Pro51.7#276
4GPT-546.7#178
5Gemini 2.5 Flash37.7#463
6Grok 410#319

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mobiflow-app-interference-completion · How It Works · Data refreshed daily, snapshot 2026-11-21.