English | 简体中文
A general-purpose foundation GUI agent for mobile apps, web platforms, and desktop operating systems — scaling environments, tasks, and feedback jointly within one closed-loop perception–reasoning–action agent.
Compared with UI-Venus-1.5, we introduce:
- 📱 Scaled multilingual mobile environments: A substantially expanded executable mobile pool covering Chinese and English app ecosystems, paired with a deep-research-driven query-generation strategy that grounds task queries in real application functionality — improving the accuracy, validity, and executability of generated instructions.
- 🖥️ Computer use, built from the ground up: Dedicated desktop operating-system capabilities constructed from scratch through computer-use data collection and task-specific training, extending the UI-Venus family to mobile, web, and OS interaction in one unified end-to-end agent.
- 🎯 Keypoint-grounded verification: Task completion is judged on task-relevant visual keypoints rather than a coarse holistic look at the final screen, with multi-model voting aggregating heterogeneous judges — reducing single-judge bias and making the reward signal robust to reward hacking.
- 🔄 Verification-augmented reflection: Verified feedback is distilled back into training as reflection supervision, so the agent can distinguish partial progress from true completion, avoid premature termination caused by observation misinterpretation, and recover during long-horizon interaction.
📈 UI-Venus-2 Benchmark Performance
Figure UI-Venus-2 on eight GUI-agent benchmarks. Each panel compares UI-Venus-2-27B and UI-Venus-2-9B with four selected strong external baselines. We favor standalone end-to-end systems evaluated on the closest available task subset and step budget; source-reported action scaffolds may still differ. Bars start at zero and scales vary by panel. MobileWorld uses GUI-only success rate on 117 tasks with 50 steps, WebVoyager uses the refreshed 595-task split, Odysseys uses average rubric score over 200 tasks, VenusBench-CAPTCHA uses micro Pass@1 over all 219 examples, and VenusBench-GD uses English-instruction micro-average accuracy. Live-site results may vary with evaluation date. “*” denotes external-model results evaluated or reproduced by us under the stated protocol, not statistical significance.
- [2026/08] We release UI-Venus-2, a 9B/27B general-purpose foundation GUI agent that unifies mobile, web, and desktop interaction with scaled multilingual environments, keypoint-grounded verification, and verification-augmented reflection.
- [2026/02] We release UI-Venus-1.5, an end-to-end GUI Agent designed for robust real-world applications.
- [2026/02] We release VenusBench-Mobile, a challenging online benchmark for mobile GUI agents. See branch VenusBench-Mobile.
- [2025/12] We release VenusBench-GD, a comprehensive multi-platform GUI grounding benchmark. See branch VenusBench-GD.
- [2025/8] We release UI-Venus, the first version of our UI agent model.
We provide two frameworks for running agents in real environments:
| Framework | Description | Documentation |
|---|---|---|
| Mobile Framework | Android/ADB agent framework for single-task execution, multi-device batch execution, trajectory recording, and reflection. | English · 中文 |
| Browser Plugin | Chrome Side Panel extension that connects UI-Venus to the active browser tab and executes browser tasks interactively. | English · 中文 |
See the Venus Framework overview for the directory layout and entry points. The lightweight domain examples below can be used without either framework unless their individual requirements state otherwise.
conda create -n ui-venus-2 python=3.11 -y
conda activate ui-venus-2
pip install -r requirements.txtPython 3.10 or newer is required. All commands below are executed from the repository root. Configure the OpenAI-compatible model service through MODEL_URL, MODEL_NAME, and either API_KEY or MODEL_API_KEY; the same values can also be edited at the top of each domain script. API_KEY takes precedence when both key variables are set. Set MODEL_NAME to the served name of your UI-Venus-2 9B or 27B model.
Run multi-turn inference over the included prerecorded screenshot sequence. N_IMG controls how many recent historical screenshots are retained:
MODEL_URL=http://127.0.0.1:8000/v1 \
MODEL_NAME=UI-Venus-2 \
N_IMG=2 \
bash scripts/mobile.shThis example performs model inference only and does not execute actions on a device. Use the Mobile Framework above for real-device ADB automation.
Mobile multi-turn example and input/output format
Run multi-turn Computer inference over a prerecorded desktop screenshot sequence:
MODEL_URL=http://127.0.0.1:8000/v1 \
MODEL_NAME=UI-Venus-2 \
N_IMG=2 \
bash scripts/computer.shThe default command uses the included desktop screenshot sample. The standalone example validates and normalizes model actions but does not execute them on the host. It has no runtime dependency on OSWorld.
Computer multi-turn example and action format
Start Chrome with a CDP port as described in the domain documentation, then run one natural-language browser task:
MODEL_URL=http://127.0.0.1:8000/v1 \
MODEL_NAME=UI-Venus-2 \
bash scripts/browser.sh "Open https://example.com and report the page title"Run the direct grounding evaluation on the three samples included with the repository:
MODEL_URL=http://127.0.0.1:8000/v1 \
MODEL_NAME=UI-Venus-2 \
bash scripts/grounding.shGrounding evaluation, smoke test, and benchmark configuration
Run inference on the included CAPTCHA image and save the parsed JSON and visualization under results/captcha/:
MODEL_URL=http://127.0.0.1:8000/v1 \
MODEL_NAME=UI-Venus-2 \
bash scripts/captcha.shCAPTCHA usage, prompt, action format, and visualization
| Models | MobileGym | VenusBench‑Mobile | AndroidWorld | MobileWorld | KnowUBench | MemGUI |
|---|---|---|---|---|---|---|
| General VLMs | ||||||
| Qwen3.5‑9B | 9.0* | 15.3* | 57.8 | 18.0 (18.0)* | 33.3 | 6.2* |
| Qwen3.6‑27B | 24.6* | 28.0* | 70.3 | 36.8 (41.9)* | - | 25.7* |
| Claude‑Opus‑4.6 | - | 36.5* | - | 44.5 | - | - |
| Kimi‑K2.6 | 38.7* | 31.2* | - | 55.6 | - | 39.1 |
| Kimi‑K3 | - | - | - | 74.4 | - | - |
| Seed‑2.0‑Pro | 52.0 | 20.1* | - | 63.2 | 51.6 | 65.6* |
| Seed‑2.1‑Pro | - | - | - | 73.2 | - | - |
| GPT‑5.6‑Sol | - | - | - | 70.1 | - | - |
| GUI‑specific Models | ||||||
| UI‑Venus‑1.5‑8B | 18.4* | 16.1 | 73.7 | 22.2* | 26.0 | 3.9* |
| UI‑Venus‑1.5‑30B‑A3B | 21.5* | 21.5 | 77.6 | 17.1 | - | 10.9* |
| GUI‑Owl‑1.5‑32B‑Instruct | 20.3* | - | 69.8 | 43.9 | - | 10.9 |
| MAI‑UI‑8B | 21.5* | 12.7 | 70.7 | 27.5 | 26.0 | 17.2* |
| Qwen‑UI‑Agent‑27B | - | - | - | 82.1 (85.5) | - | - |
| Ours | ||||||
| UI‑Venus‑2‑9B | 52.7 | 46.5 | 80.2 | 65.8 (75.2) | 56.5 | 62.6 |
| UI‑Venus‑2‑27B | 60.5 | 48.7 | 84.0 | 76.1 (82.9) | 59.7 | 70.3 |
Performance comparison on various mobile GUI benchmarks. VenusBench-Mobile reports success rate on its 149-task primary pool. For MobileWorld, we report GUI-only success rate on 117 tasks under the 50-step setting; values in parentheses, when available, use 100 steps. MemGUI reports Main Results Pass@1.
*denotes external-model results evaluated or reproduced by us; it does not denote statistical significance.
| Models | OSWorld‑Verified | OSWorld‑V2 | DeskCraft |
|---|---|---|---|
| General VLMs | |||
| Claude‑Opus‑4.8 | 83.4 | - | - |
| Qwen3.5‑9B | 41.8 | 2.5 | 14.6* |
| Qwen3.6‑27B | 62.0 | 3.8 | 28.7* |
| Kimi‑K2.6 | 73.1 | 7.1 | 41.4* |
| Seed‑2.0‑Pro | 62.3 | 6.3 | 40.0* |
| Seed‑2.1‑Pro | 78.8 | - | - |
| GPT‑5.5 | 78.7 | - | - |
| GUI‑specific Models | |||
| GUI‑Owl‑1.5‑32B‑Instruct | 56.5 | - | - |
| Qwen‑UI‑Agent‑27B | 79.5 | - | - |
| Ours | |||
| UI‑Venus‑2‑9B | 70.8 | 7.52 | 48.0 |
| UI‑Venus‑2‑27B | 80.5 | 13.24 | 55.5 |
Performance comparison on various computer-use agent benchmarks. Reported OSWorld-Verified baselines use the 361-task setting in their cited source and may use model-specific action scaffolds. For DeskCraft, we report an author-evaluated aggregate over the 538-task union of the Standard and Interactive splits; this differs from the benchmark's official split-level reporting.
*indicates external-model results evaluated by us and does not denote statistical significance.
| Models | WebVoyager | Online‑Mind2Web | REAL | Odysseys Avg. | Odysseys Perfect |
|---|---|---|---|---|---|
| General VLMs | |||||
| Qwen3.5‑9B | 46.9* | 27.3* | 18.2* | 42.6* | 13.5* |
| Qwen3.5‑4B | - | - | - | 42.9 | 10.7 |
| Qwen3.6‑27B | 84.3* | 55.3* | 27.3* | 39.5* | 18.5* |
| OpenAI Operator | 87.0 | 61.3 | - | - | - |
| GPT‑5 (SoM) | 90.6 | - | - | - | - |
| GPT‑5.4 | - | - | - | 55.4 | 33.5 |
| Seed2.0 Pro | 85.1* | 68.5* | 74.4* | 60.2* | 30.1* |
| GLM‑5V‑Turbo | 88.5 | - | - | - | - |
| Claude Opus 4.6 | 88.0 | - | - | 68.9 | 44.5 |
| Claude‑Sonnet‑4.6 | - | - | - | 49.8 | 31.0 |
| Kimi‑K2.6 | 76.8* | - | 74.4* | - | - |
| GUI‑specific Models | |||||
| UI‑TARS‑1.5 | 84.8 | 75.8 | - | - | - |
| UI‑Venus‑1.5‑30B‑A3B | 76.0 | - | 38.0* | - | - |
| GUI‑Owl‑1.5‑32B‑Thinking | 82.1 | - | 44.6* | - | - |
| MolmoWeb‑8B | 78.2 | 35.3 | - | - | - |
| Fara1.5‑4B | 80.8 | - | - | - | - |
| Fara1.5‑9B | 86.6 | 63.4 | - | - | - |
| Fara1.5‑27B | 89.3 | 72.3 | - | - | - |
| Ours | |||||
| UI‑Venus‑2‑9B | 90.8 | 74.0 | 76.9 | 77.3 | 62.0 |
| UI‑Venus‑2‑27B | 93.4 | 78.3 | 80.2 | 80.4 | 66.3 |
Performance comparison on four live-web benchmarks: WebVoyager, Online-Mind2Web, REAL, and Odysseys. The Fara1.5 and GPT-5 (SoM) WebVoyager entries use the refreshed 595-task, 100-step robust protocol and are averaged over three runs; live-site states may vary by evaluation date. For Odysseys, we report both the averaged rubric score (Avg.) and the perfect rubric score (Perfect). Bold and underlined scores indicate the best and second-best reported results in each column, respectively.
*indicates our reproduced results, which are not available from official reports.
| Models | VenusBench‑GD | ScreenSpot‑Pro | OSWorld‑G‑R | UI‑Vision |
|---|---|---|---|---|
| General VLMs | ||||
| Qwen 3.7 Plus | - | 68.9 | 78.2 | 68.0 |
| Seed 2.1 Pro | - | 65.3 | 78.0 | 62.0 |
| Kimi‑K2.6 | 73.1* | 52.0* | 69.7* | -* |
| Qwen3.6‑27B | 67.7* | 65.2* | 76.9* | 58.3* |
| GUI‑specific Models | ||||
| UI‑Venus‑Ground‑72B | 70.2 | 61.9 | 69.5 | 36.8 |
| Holo2‑30B‑A3B | 59.5* | 66.1 | 76.1 | 40.9* |
| Step‑GUI‑4B | 54.6* | 60.0 | 66.9 | 30.0* |
| MAI‑UI‑8B | 65.2* | 65.8 | 68.6 | 40.7 |
| MAI‑UI‑32B | - | 67.9 | 73.9 | 47.1 |
| UI‑Venus‑1.5‑30B‑A3B | 75.0 | 69.6 | 76.4 | 54.7 |
| Qwen‑UI‑Agent‑27B | - | 76.6 | 78.5 | 70.0 |
| Ours | ||||
| UI‑Venus‑2‑9B | 77.1 | 73.0 | 78.5 | 53.2 |
| UI‑Venus‑2‑27B | 80.1 | 74.1 | 79.1 | 66.9 |
Performance comparison on various Grounding Benchmarks. VenusBench-GD reports English-instruction micro-average point-in-box accuracy. For each benchmark, bold and underlined scores indicate the best and second-best results, respectively.
*indicates external-model results evaluated or reproduced by us under the stated protocol; it does not denote statistical significance.
| Models | VenusBench‑CAPTCHA | MCA‑Bench | Spatial‑CAPTCHA‑Bench | NextGen‑CAPTCHAs | Open CaptchaWorld |
|---|---|---|---|---|---|
| General VLMs | |||||
| Qwen3.5‑9B | 28.3* | 30.4 | 4.9 | 2.8 | 36.4 |
| Qwen3.6‑27B | 53.0* | 51.7 | 31.0 | 14.1 | 47.7 |
| Doubao‑Seed‑2.0‑Pro | 47.9* | 35.5 | 43.6 | 20.4 | 55.6 |
| Kimi‑K2.6 | 39.7* | 38.7 | 24.8 | 7.2 | 47.8 |
| Claude‑Opus‑4.6 | 16.0* | 25.9 | 9.5 | 2.8 | 23.3 |
| Ours | |||||
| UI‑Venus‑2‑9B | 78.1 | 75.7 | 42.8 | 47.6 | 50.7 |
| UI‑Venus‑2‑27B | 79.9 | 79.6 | 48.6 | 54.5 | 56.3 |
All results are Pass@1 percentages, and higher is better. VenusBench-CAPTCHA reports the micro-average over all 219 fixed examples. We evaluate on VenusBench-CAPTCHA and four public benchmarks: MCA-Bench, Spatial-CAPTCHA-Bench, NextGen-CAPTCHAs, and Open CaptchaWorld. We use 1,000 sampled MCA-Bench examples, 15 NextGen-CAPTCHAs task types, and 16 Open CaptchaWorld task types; see the appendix for selection details. Bold and underlined scores indicate the best and second-best reported results in each column, respectively. For the VenusBench-CAPTCHA column,
*denotes external-model results evaluated by us under the same protocol; it is not a significance marker.
For any questions or collaboration, please contact the maintainers.
# UI-Venus-2
@misc{venusteam2026uivenus2technicalreport,
title={UI-Venus-2 Technical Report},
author={Venus-Team and xxx},
year={2026},
eprint={xxx},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={comming soon},
}
# UI-Venus-1.5
@misc{venusteam2026uivenus15technicalreport,
title={UI-Venus-1.5 Technical Report},
author={Venus-Team and Changlong Gao and Zhangxuan Gu and Yulin Liu and Xinyu Qiu and Shuheng Shen and Yue Wen and Tianyu Xia and Zhenyu Xu and Zhengwen Zeng and Beitong Zhou and Xingran Zhou and Weizhi Chen and Sunhao Dai and Jingya Dou and Yichen Gong and Yuan Guo and Zhenlin Guo and Feng Li and Qian Li and Jinzhen Lin and Yuqi Zhou and Linchao Zhu and Liang Chen and Zhenyu Guo and Changhua Meng and Weiqiang Wang},
year={2026},
eprint={2602.09082},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2602.09082},
}
# UI-Venus
@misc{gu2025uivenustechnicalreportbuilding,
title={UI-Venus Technical Report: Building High-performance UI Agents with RFT},
author={Zhangxuan Gu and Zhengwen Zeng and Zhenyu Xu and others},
year={2025},
eprint={2508.10833},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2508.10833},
}This project is for research and educational purposes only.