OmniGUI
Models use screenshots, sound, and video to predict the next action in 29 phone apps. The grader compares each action with a human demonstration.
XPeng ranks nine models on 708 episodes with 2,572 actions after filtering invalid samples.
- Made by
- XPeng Motors
- Released
- March 2026
- Kind
- Phones
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Interaction
- Single turn
- Items from
- Written by experts, Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Audio and speech
- Strongest model tested
-
Gemini 3.1 Pro #40 on the capability index
- Last model added
- February 2026
- Scores refreshed
- November 20th, 2026
Gemini 3.1 Pro leads with 65.4% among models released by February 2026 Gemini 3.1 Pro leads with 65.4% among models released by February 2026
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 February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
Gemini 3.1 Pro 65.4%
- 2
Gemini 3 Pro 63.7%
- 3
Gemini 3 Flash 61.4%
- 4
Gemini 2.5 Pro Jun 2025 44.1%
- 5
Gemini 2.5 Flash Sep 2025 37.8%
- 6 Qwen3-Omni 32.4%
- 7 VITA-1.5 12.1%
- 8
MiniCPM-o 4.5 4.90%
- 9
Baichuan-Omni-1.5 3.30%
Source: Official OmniGUI leaderboard
- 1
Gemini 3.1 Pro 65.4%
- 2
Gemini 3 Pro 63.7%
- 3
Gemini 3 Flash 61.4%
- 4
Gemini 2.5 Pro Jun 2025 44.1%
- 5
Gemini 2.5 Flash Sep 2025 37.8%
- 6 VITA-1.5 12.1%
- 7
MiniCPM-o 4.5 4.90%
- 8
Baichuan-Omni-1.5 3.30%
Source: Official OmniGUI leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 65.4% | Feb 2026 | Official leaderboard | |
| 2 | | 63.7% | Nov 2025 | Official leaderboard | |
| 3 | | 61.4% | Dec 2025 | Official leaderboard | |
| 4 | | 44.1% | Jun 2025 | Official leaderboard | |
| 5 | | 37.8% | Sep 2025 | Official leaderboard | |
| 6 | Qwen3-Omni | 32.4% | Official leaderboard | ||
| 7 | VITA-1.5 | 12.1% | Dec 2024 | Official leaderboard | |
| 8 | | 4.90% | Feb 2026 | Official leaderboard | |
| 9 | | 3.30% | Jan 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 65.4% | Feb 2026 | Official leaderboard | |
| 2 | | 63.7% | Nov 2025 | Official leaderboard | |
| 3 | | 61.4% | Dec 2025 | Official leaderboard | |
| 4 | | 44.1% | Jun 2025 | Official leaderboard | |
| 5 | | 37.8% | Sep 2025 | Official leaderboard | |
| 6 | VITA-1.5 | 12.1% | Dec 2024 | Official leaderboard | |
| 7 | | 4.90% | Feb 2026 | Official leaderboard | |
| 8 | | 3.30% | Jan 2025 | Official leaderboard |
Rankings by part
We read 24 parts from the benchmark’s own leaderboard in 6 groups: localization, semantic understanding, cross-modal discrimination, temporal reasoning, instant response and overall. Each ranks models on one part of the test. Gemini 3.1 Pro leads 20 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 87.9%
- 2
Gemini 3 Pro 86.3%
- 3
Gemini 2.5 Pro Jun 2025 ↑1 86.1%
- 4
Gemini 3 Flash ↓1 85%
- 5
Gemini 2.5 Flash Sep 2025 75.1%
- 6 VITA-1.5 48.4%
- 7
MiniCPM-o 4.5 34.8%
- 8
Baichuan-Omni-1.5 19.5%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 77.8%
- 2
Gemini 3 Pro 76.2%
- 3
Gemini 3 Flash 75.6%
- 4
Gemini 2.5 Pro Jun 2025 58.1%
- 5
Gemini 2.5 Flash Sep 2025 50.9%
- 6 VITA-1.5 14.8%
- 7
MiniCPM-o 4.5 7.40%
- 8
Baichuan-Omni-1.5 4.90%
Success Rate
8 models we track have a score here.
- 1
Gemini 3.1 Pro 57.9%
- 2
Gemini 3 Pro 55.9%
- 3
Gemini 3 Flash 53.1%
- 4
Gemini 2.5 Pro Jun 2025 31.7%
- 5
Gemini 2.5 Flash Sep 2025 29%
- 6 VITA-1.5 2.80%
- 7
MiniCPM-o 4.5 0.70%
- 8
Baichuan-Omni-1.5 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3.1 Pro 66.2%
- 2
Gemini 3 Flash ↑1 63.1%
- 3
Gemini 3 Pro ↓1 62.6%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 42.6%
- 5
Gemini 2.5 Pro Jun 2025 ↓1 41.5%
- 6 VITA-1.5 3.90%
- 7
MiniCPM-o 4.5 2.20%
- 8
Baichuan-Omni-1.5 1.00%
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 80%
- 2
Gemini 3 Pro 77.4%
- 3
Gemini 3 Flash 75.3%
- 4
Gemini 2.5 Pro Jun 2025 72.8%
- 5
Gemini 2.5 Flash Sep 2025 70.4%
- 6 VITA-1.5 43.4%
- 7
MiniCPM-o 4.5 34.7%
- 8
Baichuan-Omni-1.5 16.2%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 61.9%
- 2
Gemini 3 Pro 61.1%
- 3
Gemini 3 Flash 58.5%
- 4
Gemini 2.5 Pro Jun 2025 37.7%
- 5
Gemini 2.5 Flash Sep 2025 34.3%
- 6 VITA-1.5 16.4%
- 7
MiniCPM-o 4.5 5.50%
- 8
Baichuan-Omni-1.5 4.00%
Success Rate
8 models we track have a score here.
- 1
Gemini 3 Pro ↑1 31.4%
- 2
Gemini 3.1 Pro ↓1 31.4%
- 3
Gemini 3 Flash 25.5%
- 4
Gemini 2.5 Pro Jun 2025 11.7%
- 5
Gemini 2.5 Flash Sep 2025 8.00%
- 6 VITA-1.5 2.20%
- 7
Baichuan-Omni-1.5 ↑1 0%
- 8
MiniCPM-o 4.5 ↓1 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3 Pro ↑1 42%
- 2
Gemini 3.1 Pro ↓1 41.4%
- 3
Gemini 3 Flash 41.1%
- 4
Gemini 2.5 Pro Jun 2025 22.4%
- 5
Gemini 2.5 Flash Sep 2025 18.2%
- 6 VITA-1.5 3.20%
- 7
MiniCPM-o 4.5 1.00%
- 8
Baichuan-Omni-1.5 0.50%
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 81.1%
- 2
Gemini 3 Pro 76.8%
- 3
Gemini 3 Flash 73.2%
- 4
Gemini 2.5 Pro Jun 2025 70.6%
- 5
Gemini 2.5 Flash Sep 2025 64.8%
- 6 VITA-1.5 33.5%
- 7
MiniCPM-o 4.5 25.9%
- 8
Baichuan-Omni-1.5 12.6%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 63%
- 2
Gemini 3 Pro 59.3%
- 3
Gemini 3 Flash 56.4%
- 4
Gemini 2.5 Pro Jun 2025 39.5%
- 5
Gemini 2.5 Flash Sep 2025 35.6%
- 6 VITA-1.5 11.7%
- 7
MiniCPM-o 4.5 4.10%
- 8
Baichuan-Omni-1.5 1.40%
Success Rate
8 models we track have a score here.
- 1
Gemini 3.1 Pro 38.3%
- 2
Gemini 3 Pro 29.8%
- 3
Gemini 3 Flash 23.4%
- 4
Gemini 2.5 Pro Jun 2025 12.8%
- 5
Gemini 2.5 Flash Sep 2025 11.3%
- 6 VITA-1.5 0%
- 7
Baichuan-Omni-1.5 ↑1 0%
- 8
MiniCPM-o 4.5 ↓1 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3.1 Pro 46.1%
- 2
Gemini 3 Pro 41.2%
- 3
Gemini 3 Flash 39.1%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 24.9%
- 5
Gemini 2.5 Pro Jun 2025 ↓1 24.5%
- 6
MiniCPM-o 4.5 ↑1 1.80%
- 7 VITA-1.5 ↓1 1.00%
- 8
Baichuan-Omni-1.5 0%
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 82.9%
- 2
Gemini 3 Flash ↑1 80.4%
- 3
Gemini 3 Pro ↓1 79.1%
- 4
Gemini 2.5 Pro Jun 2025 73.9%
- 5
Gemini 2.5 Flash Sep 2025 68%
- 6 VITA-1.5 35.9%
- 7
MiniCPM-o 4.5 35.7%
- 8
Baichuan-Omni-1.5 18.9%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 61.8%
- 2
Gemini 3 Pro 60.7%
- 3
Gemini 3 Flash 60%
- 4
Gemini 2.5 Pro Jun 2025 44.1%
- 5
Gemini 2.5 Flash Sep 2025 35.1%
- 6 VITA-1.5 7.70%
- 7
MiniCPM-o 4.5 4.20%
- 8
Baichuan-Omni-1.5 2.40%
Success Rate
8 models we track have a score here.
- 1
Gemini 3 Flash ↑2 25.6%
- 2
Gemini 3.1 Pro ↓1 24.4%
- 3
Gemini 3 Pro ↓1 23.1%
- 4
Gemini 2.5 Pro Jun 2025 10.3%
- 5
Gemini 2.5 Flash Sep 2025 9.00%
- 6 VITA-1.5 0.60%
- 7
MiniCPM-o 4.5 0.60%
- 8
Baichuan-Omni-1.5 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3 Flash ↑2 38.6%
- 2
Gemini 3.1 Pro ↓1 36.4%
- 3
Gemini 3 Pro ↓1 36.2%
- 4
Gemini 2.5 Pro Jun 2025 22.5%
- 5
Gemini 2.5 Flash Sep 2025 21.1%
- 6 VITA-1.5 1.90%
- 7
MiniCPM-o 4.5 1.20%
- 8
Baichuan-Omni-1.5 0.20%
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 85.7%
- 2
Gemini 3 Pro 81.4%
- 3
Gemini 3 Flash 78.7%
- 4
Gemini 2.5 Pro Jun 2025 77.2%
- 5
Gemini 2.5 Flash Sep 2025 71.2%
- 6 VITA-1.5 37.1%
- 7
MiniCPM-o 4.5 33%
- 8
Baichuan-Omni-1.5 17.9%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 64.8%
- 2
Gemini 3 Pro 63.1%
- 3
Gemini 3 Flash 58.4%
- 4
Gemini 2.5 Pro Jun 2025 43.1%
- 5
Gemini 2.5 Flash Sep 2025 35.4%
- 6 VITA-1.5 10.9%
- 7
Baichuan-Omni-1.5 ↑1 4.10%
- 8
MiniCPM-o 4.5 ↓1 3.40%
Success Rate
8 models we track have a score here.
- 1
Gemini 3.1 Pro 31.8%
- 2
Gemini 3 Pro 27.1%
- 3
Gemini 3 Flash 23.3%
- 4
Gemini 2.5 Pro Jun 2025 11.6%
- 5
Gemini 2.5 Flash Sep 2025 4.70%
- 6 VITA-1.5 0%
- 7
Baichuan-Omni-1.5 ↑1 0%
- 8
MiniCPM-o 4.5 ↓1 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3.1 Pro 40.3%
- 2
Gemini 3 Pro 35.7%
- 3
Gemini 3 Flash 34.8%
- 4
Gemini 2.5 Pro Jun 2025 20%
- 5
Gemini 2.5 Flash Sep 2025 14.8%
- 6
MiniCPM-o 4.5 ↑1 1.00%
- 7 VITA-1.5 ↓1 0.80%
- 8
Baichuan-Omni-1.5 0.50%
Type Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 83.3%
- 2
Gemini 3 Pro 80%
- 3
Gemini 3 Flash 78.4%
- 4
Gemini 2.5 Pro Jun 2025 75.7%
- 5
Gemini 2.5 Flash Sep 2025 69.7%
- 6 VITA-1.5 39.4%
- 7
MiniCPM-o 4.5 32.9%
- 8
Baichuan-Omni-1.5 17%
Exact Match
8 models we track have a score here.
- 1
Gemini 3.1 Pro 65.4%
- 2
Gemini 3 Pro 63.7%
- 3
Gemini 3 Flash 61.4%
- 4
Gemini 2.5 Pro Jun 2025 44.1%
- 5
Gemini 2.5 Flash Sep 2025 37.8%
- 6 VITA-1.5 12.1%
- 7
MiniCPM-o 4.5 4.90%
- 8
Baichuan-Omni-1.5 3.30%
Success Rate
8 models we track have a score here.
- 1
Gemini 3.1 Pro 36.7%
- 2
Gemini 3 Pro 33.5%
- 3
Gemini 3 Flash 30.4%
- 4
Gemini 2.5 Pro Jun 2025 15.7%
- 5
Gemini 2.5 Flash Sep 2025 12.6%
- 6 VITA-1.5 1.10%
- 7
MiniCPM-o 4.5 0.30%
- 8
Baichuan-Omni-1.5 0%
Goal Progress
8 models we track have a score here.
- 1
Gemini 3.1 Pro 46.1%
- 2
Gemini 3 Pro 43.6%
- 3
Gemini 3 Flash 43.5%
- 4
Gemini 2.5 Pro Jun 2025 26.3%
- 5
Gemini 2.5 Flash Sep 2025 24.6%
- 6 VITA-1.5 2.20%
- 7
MiniCPM-o 4.5 1.50%
- 8
Baichuan-Omni-1.5 0.40%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official OmniGUI 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 OmniGUI 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 9 months ago, in February 2026, for Gemini 3.1 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 (9 entries).
- Pass Models scored 8 models we track, plus 1 other entry on its leaderboard.
- Warning Newest score The newest came 9 months ago, in February 2026, for Gemini 3.1 Pro.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.1 Pro (#40).
- Pass Headroom The best is 65.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on omni-gui.github.io: 8 rows for models we track and 1 other entry 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.