CAGUI
Models locate controls and read marked text in Chinese Android screenshots. A separate test checks each predicted action against the recorded steps for an app task.
The DKR-GUI team ran the action test. The GGA team ran the two grounding tasks that ask models to locate controls.
- Made by
- OpenBMB
- Released
- May 2025
- Kind
- Phones
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- Chinese
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Last model added
- January 2025
- Scores refreshed
- November 20th, 2026
AgentCPM-GUI-8B tops the leaderboard with 84.1% Qwen2.5-VL-7B leads with 65.3%
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The authors report no overall score, so this is the plain mean of grounding average and action type match. 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 Qwen2.5-VL-7B, at 65.3%. This view shows only models we track, each at its best.
- 1 AgentCPM-GUI-8B 84.1%
- 2
Qwen2.5-VL-7B 65.3%
- 3 UI-TARS-7B 65.1%
- 4 OS-Atlas-7B 59.9%
- 5 Aguvis-7B 56.6%
- 6 OS-Genesis-7B 22.1%
- 7 GPT-4o (snapshot unspecified) 11.2%
Source: Official CAGUI leaderboard
- 1
Qwen2.5-VL-7B 65.3%
Source: Official CAGUI leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | AgentCPM-GUI-8B | 84.1% | Official leaderboard | ||
| 2 | | 65.3% | Jan 2025 | Official leaderboard | |
| 3 | UI-TARS-7B | 65.1% | Official leaderboard | ||
| 4 | OS-Atlas-7B | 59.9% | Official leaderboard | ||
| 5 | Aguvis-7B | 56.6% | Official leaderboard | ||
| 6 | OS-Genesis-7B | 22.1% | Official leaderboard | ||
| 7 | GPT-4o (snapshot unspecified) | 11.2% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 65.3% | Jan 2025 | 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 CAGUI 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 22 months ago, in January 2025, for Qwen2.5-VL-7B. 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 (7 entries).
- Pass Models scored 1 model we track, plus 6 other entries on its leaderboard.
- Warning Newest score The newest came 22 months ago, in January 2025, for Qwen2.5-VL-7B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 65.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 1 row for models we track and 6 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.