@AliceInfoAi: 100% on Google’s AndroidWorld benchmark. Mobilerun completed all 116/116 tasks across 20 Android apps and multi step wo…
Summary
MobileRun claims a perfect 116/116 score on Google's AndroidWorld benchmark, covering 20 Android apps and multi-step workflows, reportedly outperforming Google's Artemis agent at 99%+.
View Cached Full Text
Cached at: 10/02/26, 10:39 AM
100% on Google’s AndroidWorld benchmark.
Mobilerun completed all 116/116 tasks across 20 Android apps and multi step workflows.
That’s a serious milestone for AI agents operating on Android.
#Mobilerun #AI #Android
Niels Schmidt (@nielsmdt99): Breaking: Mobilerun just beat GOOGLE’S ARTEMIS on AndroidWorld🔥
We ran the full benchmark and mobilerun completed 116/116 tasks with a full official score on Google’s AndroidWorld benchmark, spanning 20 Android apps and multi-step workflows.
Google Artemis scored 99%+ and
Similar Articles
@yoheinakajima: just ran this on my phone (17 pro)
RunAnywhere releases an on-device AI benchmark app that lets users measure model performance directly on their own phones, replacing posted numbers with real-time testing.
AndroidLife: Can an AI agent survive a day in the life of a real user? Qwen3.8-27b run: 56.7% SR
An AI agent using the Qwen3.8-27b model from Alibaba achieved a 56.7% success rate in the AndroidLife benchmark, running 60 real tasks on a OnePlus phone with detailed performance and efficiency metrics.
@googledevs: See how AI models can help you with multi-day engineering workflows with Android Bench 2.0. The updated benchmark evalu…
Android Bench 2.0 is an updated AI evaluation framework that assesses models on long-horizon engineering tasks for Android, such as building apps from scratch and migrating codebases, with continuous scoring.
SimuWoB: Simulating Real-World Mobile Apps for Fast and Faithful GUI Agent Benchmarking
SimuWoB is a synthetic benchmark with 120 challenging tasks for mobile GUI agents, using high-fidelity virtual environments and automatic reward generation. Experiments reveal that current agents achieve only 27.92% average success rate, dropping to 17.82% on long-horizon tasks, indicating substantial weaknesses in complex scenarios.
Google updates Android Bench with new LLMs, but Gemini still lags behind
Google updates Android Bench, a benchmark for LLMs in Android development tasks, adding new models like Claude Fable 5 and Qwen 3.7 Max, but its own Gemini models still trail behind competitors.