LLM Reference
activeAgents

AndroidWorld

Metric: Accuracy (higher is better)Introduced: 2024

Android mobile-agent benchmark for measuring task completion and GUI automation on Android environments. High benchmark score alone doesn't make a model the right pick — weigh it against pricing, API availability, and release date.

Models ranked

4

tracked on this benchmark

Score band

79.3 – 71.0

best → lowest tracked

Snapshot trend

+2.17

Nov 13 → Jun 1 · 3 models

Leaderboard

Tracked models ranked by Accuracy (higher is better).

Compare candidates
#Model variant and provenanceScore
1
Holo3.1-35B-A3B
Version: AndroidWorld - mobile agentHarness: Not recordedEvaluator: Not recordedObserved: Jun 1, 2026Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

79.3
2
Holo2-30B-A3B
Version: AndroidWorld - mobile agentHarness: Not recordedEvaluator: Not recordedObserved: Nov 13, 2025Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

71.6
3
Holo3.1-4B
Version: AndroidWorld - mobile agentHarness: Not recordedEvaluator: Not recordedObserved: Jun 1, 2026Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

71.0
4
Holo3.1-9B
Version: AndroidWorld - mobile agentHarness: Not recordedEvaluator: Not recordedObserved: Jun 1, 2026Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

71.0

How to read this benchmark

This benchmark scores models where higher is better. Scores are useful for directional filtering and shortlisting — not for universal quality ranking. Prefer benchmarks closest to your workload, then validate the linked model pages for pricing, context window, and provider availability.

Trust this score when

  • There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
  • The model list covers the same version family you can actually deploy today.
  • Top candidates overlap with your required routing and feature requirements.

Be cautious when

  • There is only one benchmark snapshot or the dataset appears stale.
  • The benchmark metric direction is opposite of your decision objective.
  • The score difference between options is narrow and likely within implementation variance.

FAQ

What does the AndroidWorld benchmark measure?

Android mobile-agent benchmark for measuring task completion and GUI automation on Android environments. On this page it lists 4 tracked model variants where higher is better.

Is a higher AndroidWorld score always better?

For this benchmark, higher is better. A high score helps you shortlist, but confirm pricing, context window, and provider availability on each model page before committing — the top scorer is not always the right pick for your workload or budget.

How current is this AndroidWorld data?

This benchmark was last reviewed on Jun 7, 2026. The tracked score average moved +2.17 points across the last 2 snapshots.

Related benchmarks

Last reviewed: Jun 7, 2026

Resources