Benchmark

AndroidWorld

AndroidWorld evaluates an agent's ability to operate in real Android GUI environments, completing multi-step tasks by perceiving screen content and…

Modality
multimodal
Categories
agents, vision
Openness
unknown
Source
llm_stats
Reported scores
5

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
GLM-5V-TurboZ.ai0.7572026-04-02evidence
Qwen3.6-27BQwen0.7032026-04-21evidence
Qwen3.7-PlusQwen0.812026-05-31evidence
Qwen3.8 MaxQwen0.8532026-08-02evidence
Qwen3.8-27BQwen0.8192026-08-14evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.