MobileWorld - MCP: leaderboard
The MCP split of MobileWorld: 40 tasks where the agent must combine GUI control with MCP tool calls; success rate %, end-to-end models only.
Metric: Success Rate (%). Source: tongyi-mai.github.io. Saturation forecast: Around 2029. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B | 5.4 | #251 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mobileworld-mcp · How It Works · Data refreshed daily, snapshot 2026-11-05.