Benchmark Almanac

Android Agent Arena (100 tasks)

Gemini 2.5 Pro Jun 2025 53% leads since August 2025
Not in model cards
Version 100 tasks

Agents use 20 Android apps to find information and perform actions. A vision model checks whether the agent reached the required intermediate states and completed the task.

The authors rank ten agent configurations on 100 tasks. They report final success and the percentage of required intermediate states each agent reached.

Released
August 2025
Kind
Phones
Level
Everyday
Scored
Graded by a model
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Screens → Actions
Interaction
Agentic
Environment
Phone apps
Run by
Anyone
In views
Computer use, Agents
Strongest model tested
Gemini 2.5 Pro Jun 2025 #97 on the capability index
Last model added
June 2025
Scores refreshed
November 20th, 2026
8
Entries on its own leaderboard, 3 models we track

Gemini 2.5 Pro leads with 53% among models released by June 2025 Gemini 2.5 Pro leads with 53% among models released by June 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

The newest model with a score here came out in June 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 2.5 Pro Jun 2025 53%
  2. 2 InfiGUI-R1 27%
  3. 3 UI-Venus 20%
  4. 4 Qwen3-VL 17%
  5. 5 Qwen2.5-VL-7B 16%
  6. 6 GUI-OWL 14%
  7. 7 UI-Genie 13%
  8. 8 UI-TARS-1.5 12%

Source: Official Android Agent Arena (100 tasks) leaderboard

Source: Official Android Agent Arena (100 tasks) leaderboard

#EntryScoreSettingDateSource
1 Gemini 2.5 Pro Jun 2025 53% T3A Jun 2025 Official leaderboard
2 InfiGUI-R1 7B 27% Official leaderboard
3 UI-Venus 7B 20% Official leaderboard
4 Qwen3-VL 7B 17% Official leaderboard
5 Qwen2.5-VL-7B 16% Mobile-Use Jan 2025 Official leaderboard
6 GUI-OWL 7B 14% Official leaderboard
7 UI-Genie 7B 13% Official leaderboard
8 UI-TARS-1.5 12% Apr 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 2.5 Pro Jun 2025 53% T3A Jun 2025 Official leaderboard
2 Qwen2.5-VL-7B 16% Mobile-Use Jan 2025 Official leaderboard
3 UI-TARS-1.5 12% Apr 2025 Official leaderboard

Rankings by part

We read 10 parts from the benchmark’s own leaderboard in 2 groups: success rate (sr) and essential state achieved rate (esar). Each ranks models on one part of the test. Gemini 2.5 Pro leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Easy

3 models we track have a score here.

  1. 1 Gemini 2.5 Pro Jun 2025 57.1%
  2. 2 Qwen2.5-VL-7B 34.3%
  3. 3 UI-TARS-1.5 20%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Own leaderboard Not tested since Aug 2025

Source: Official Android Agent Arena (100 tasks) leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Gemini 2.5 Pro Qwen2.5-VL-7B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Gemini 2.5 Pro Qwen2.5-VL-7B

Source: Official Android Agent Arena (100 tasks) leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

More in Computer and browser use

All 119 in Computer and browser use
  • BrowseComp In use Hard-to-find facts that need persistent web browsing. Short verifiable answers graded for exact correctness. Claude Fable 5 93.3%
  • ScreenSpot-Pro In use GUI grounding on high-resolution professional-software screenshots. The model must locate the target UI element. GPT-6 Astra 92.7%
  • AndroidWorld In use Agents complete 116 tasks across 20 real Android apps in an emulator. Each task checks the device's system state to judge success. GPT-6 Astra 100%
  • OSWorld 2.1 In use Agents complete 108 long-horizon workflows across 31 self-hosted websites and desktop apps, averaging 318 tool calls. Scored by binary completion and checkpoint partial credit. Claude Opus 5 44.3%
  • WebArena-Verified In use Agents complete 812 audited WebArena tasks on self-hosted websites. Deterministic evaluators check the agent's answer and captured network traces. Muse Spark 1.1 69%
  • OSWorld 2.1 (offline subset) In use Agents complete the 82 OSWorld 2.1 tasks on the official offline list, with no internet access. Scored by checkpoint partial credit and strict pass rate. Claude Fable 5.2 89.3%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 17 months ago, in June 2025, for Gemini 2.5 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (8 entries).
  • Pass Models scored 3 models we track, plus 5 other entries on its leaderboard.
  • Warning Newest score The newest came 17 months ago, in June 2025, for Gemini 2.5 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 2.5 Pro (#97).
  • Pass Headroom The best is 53%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on yxchai.com: 5 rows for models we track and 5 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check