AndroidArena
Agents use apps such as Gmail and Google Maps, sometimes switching apps or following extra constraints. GPT-4 judges success from the recorded actions and observations.
MobileJudgeBench released 151 trajectories from 51 AndroidArena tasks. The original paper counts 164 single-app tasks, 22 cross-app tasks, and 35 constrained tasks.
- Made by
- Peking University and Microsoft Research
- Released
- February 2024
- Kind
- Phones
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Phone apps
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Computer use, Agents
- Strongest model tested
-
Llama 2-70B #210 on the capability index
- Last model added
- July 2023
- Scores refreshed
- November 20th, 2026
GPT-4 (snapshot unspecified) tops the leaderboard with 75.9% Llama 2-70B leads with 12.7%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The benchmark reports separate completion and constraint metrics, so this is the single-app task success rate. 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. The best model we track is Llama 2-70B, at 12.7%. This view shows only models we track, each at its best.
- 1 GPT-4 (snapshot unspecified) 75.9%
- 2 GPT-3.5 (snapshot unspecified) 44.9%
- 3
Llama 2-70B 12.7%
- 4
Llama 2-13B 2.30%
Source: The AndroidArena paper
- 1
Llama 2-70B 12.7%
- 2
Llama 2-13B 2.30%
Source: The AndroidArena paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4 (snapshot unspecified) | 75.9% | Official leaderboard | ||
| 2 | GPT-3.5 (snapshot unspecified) | 44.9% | Official leaderboard | ||
| 3 | | 12.7% | Jul 2023 | Official leaderboard | |
| 4 | | 2.30% | Jul 2023 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 12.7% | Jul 2023 | Official leaderboard | |
| 2 | | 2.30% | Jul 2023 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official AndroidArena 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.
More 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
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in July 2023, for Llama 2-13B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (4 entries).
- Note Models scored Only 2 models we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 3 years ago, in July 2023, for Llama 2-13B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Llama 2-70B (#210).
- Pass Headroom The best is 12.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 2 rows for models we track and 2 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.