AndroidWorld
Progress Over Time
Interactive timeline showing model performance evolution on AndroidWorld
State-of-the-art frontier
Open
Proprietary
AndroidWorld Leaderboard
6 models
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 2.4T | 1.0M | $1.65 / $4.95 | ||
| 2 | Alibaba Cloud / Qwen Team | 125B | — | — | ||
| 3 | Alibaba Cloud / Qwen Team | 28B | — | — | ||
| 4 | Alibaba Cloud / Qwen Team | — | — | — | ||
| 5 |  Zhipu AI | — | — | — | ||
| 6 | Alibaba Cloud / Qwen Team | 28B | 262K | $0.60 / $3.60 |
Notice missing or incorrect data?
What is AndroidWorld?
AndroidWorld evaluates an agent's ability to operate in real Android GUI environments, completing multi-step tasks by perceiving screen content and executing touch/type actions.
AndroidWorld is a multimodal benchmark evaluating models on agents and vision tasks. LLM Stats tracks 6 models on this benchmark, scored on a 0–1 scale. The current average is 0.8, with the leader at 0.9.
Compare leaders on the best AI for agents and best AI for vision leaderboards.
Current leaders
Qwen3.8 Max from Alibaba Cloud / Qwen Team currently leads the AndroidWorld leaderboard with a score of 0.853 across 6 evaluated AI models.
FAQ
Common questions about the AndroidWorld benchmark and leaderboard.