PhoneHarness Bench
Agents complete Android workflows through graphical controls, device commands, and structured tools. Checks inspect files, system settings, application state, and unintended side effects.
PhoneHarness Bench is the evaluation dataset. PhoneHarness is also the name of the surrounding execution framework.
- Made by
- Tencent Hunyuan, CUHK, CUHK-Shenzhen and others
- Released
- June 2026
- Kind
- Phones
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Phone apps, Terminal, APIs and tools
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Computer use, Agents
- Last model added
- February 2026
- Scores refreshed
- November 18th, 2026
In the authors’ words
Phone agents are increasingly expected to complete real mobile workflows rather than merely predict the next screen action. However, much of the current mobile-agent literature still evaluates agents primarily as GUI controllers that observe a screen, emit taps and swipes, and are scored by target app state. Real phone-use tasks are broader: they require deciding when to use app GUIs, device-side commands, or structured tools, while leaving evidence that the intended side effect actually occurred. We introduce PhoneHarness, a mixed-action benchmark and execution harness for studying phone-use agents on verifiable mobile workflows. PhoneHarness runs a device-side agent loop over GUI, CLI, and host-side tool actions, combining deterministic action routing with bounded GUI delegation and auditable execution traces. Its benchmark, PhoneHarness Bench, evaluates whether agents complete tasks with observable side effects, not only whether they produce plausible final answers. On the annotated evaluation split, PhoneHarness reaches a 75.0% pass rate, outperforming the strongest non-PhoneHarness settings by 12.9 percentage points. PhoneHarness and PhoneHarness Bench therefore play distinct but mutually dependent roles: the harness makes mixed phone workflows executable, while the benchmark measures whether agents can use that harness reliably and safely. Our findings suggest that reliable phone automation depends on action-surface routing and verifiable execution, not only visual GUI control.
PhoneHarness (backbone unspecified) tops the leaderboard with 75% Seed2.0 Pro leads with 62.1%
Best score per entry from the benchmark's paper, read on November 18th, 2026. A paper’s table doesn’t add newer models. 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 Seed2.0 Pro, at 62.1%. This view shows only models we track, each at its best.
- 1 PhoneHarness (backbone unspecified) 75%
- 2 MobileClaw (backbone unspecified) 62.1%
- 3
Seed2.0 Pro 62.1%
- 4 AutoGLM-Phone 37.1%
Source: The PhoneHarness Bench paper
- 1
Seed2.0 Pro 62.1%
Source: The PhoneHarness Bench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | PhoneHarness (backbone unspecified) | 75% | PhoneHarness | Official leaderboard | |
| 2 | MobileClaw (backbone unspecified) | 62.1% | MobileClaw | Official leaderboard | |
| 3 | | 62.1% | Feb 2026 | Official leaderboard | |
| 4 | AutoGLM-Phone | 37.1% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 62.1% | Feb 2026 | 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 PhoneHarness Bench 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 9 months ago, in February 2026, for Seed2.0 Pro. 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 1 model we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 9 months ago, in February 2026, for Seed2.0 Pro.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 62.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 1 row for models we track and 3 other entries at the last read.
- Pass Last read November 18th, 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.