# MobileWorld

> A mobile-use agent benchmark for completing interactive tasks in smartphone environments.

Canonical page: https://benchlm.ai/benchmarks/mobileworld

- Category: [Agentic](/agentic)
- Last updated: September 27, 2026

## About MobileWorld

- Year: 2026
- Tasks: Interactive mobile-device workflows
- Format: Mobile agent task score
- Difficulty: Long-horizon mobile computer use
- Paper: [Qwen3.8-Max: A New Bar for Coding and Cowork](https://qwen.ai/blog?id=qwen3.8)

Qwen reports MobileWorld in its visual-agent comparison table. We keep the provider-run result display-only while the public protocol and cross-provider harness coverage remain sparse.

MobileWorld is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (1 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Qwen3.8 Max](/models/qwen3-8-max) | Alibaba | 77.8% |

## FAQ

### What does MobileWorld measure?

A mobile-use agent benchmark for completing interactive tasks in smartphone environments.

### Which model scores highest on MobileWorld?

Qwen3.8 Max by Alibaba currently leads with a score of 77.8% on MobileWorld.

### How many models are evaluated on MobileWorld?

1 AI models have been evaluated on MobileWorld on BenchLM.

### Does MobileWorld affect BenchLM's overall score?

Not directly. MobileWorld is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.