---
title: MAI-UI-2B
canonical_url: "https://www.modelscope.cn/models/Tongyi-MAI/MAI-UI-2B"
md_url: "https://www.modelscope.cn/models/Tongyi-MAI/MAI-UI-2B.md"
repository: Tongyi-MAI/MAI-UI-2B
last_updated: 2026-01-04
license: apache-2.0
model_type:
  - qwen3_vl
architectures:
  - Qwen3VLForConditionalGeneration
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 1055
stars: 7
---

# MAI-UI-2B

> MAI-UI-2B - Tongyi-MAI 在 ModelScope 开源的模型。MAI-UI: Real-World Centric Foundation GUI Agents.

Tongyi-MAI/MAI-UI-2B 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: Tongyi-MAI/MAI-UI-2B
- **License**: apache-2.0
- **Downloads**: 1055
- **Stars**: 7
- **Last updated**: 2026-01-04

Source: https://www.modelscope.cn/models/Tongyi-MAI/MAI-UI-2B

---

# MAI-UI: Real-World Centric Foundation GUI Agents.
![overview](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/_ibfeHy_31ZRanQ3xxlnn.png)

## 📖 Background
The development of GUI agents could revolutionize the next generation of human-computer interaction. Motivated by this vision, we present MAI-UI, a family of foundation GUI agents spanning the full spectrum of sizes, including 2B, 8B, 32B, and 235B-A22B variants. We identify four key challenges to realistic deployment: the lack of native agent–user interaction, the limits of UI-only operation, the absence of a practical deployment architecture, and brittleness in dynamic environments. MAI-UI addresses these issues with a unified methodology: a self-evolving data pipeline that expands the navigation data to include user interaction and MCP tool calls, a native device–cloud collaboration system that routes execution by task state, and an online RL framework with advanced optimizations to scale parallel environments and context length. 

## 🏆 Results

### Grounding
MAI-UI establishes new state-of-the-art across GUI grounding and mobile navigation. 

- On grounding benchmarks, it reaches 73.5% on ScreenSpot-Pro, 91.3% on MMBench GUI L2, 70.9% on OSWorld-G, and 49.2% on UI-Vision, surpassing Gemini-3-Pro and Seed1.8 on ScreenSpot-Pro. 
![sspro](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/oNggS-XoVZSiiLitn8nD3.jpeg)
![uivision](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/8t_oRozJ51aw5rKojqlOu.jpeg)
![mmbench](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/dwc8np4JLxTWoNS8wu18p.jpeg)
![osworld-g](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/lff2k7ZkMgCrgt3lPZChb.jpeg)

### Mobile Navigation
- On mobile GUI navigation, it sets a new SOTA of 76.7% on AndroidWorld, surpassing UI-Tars-2, Gemini-2.5-Pro and Seed1.8. On MobileWorld, MAI-UI obtains 41.7% success rate, significantly outperforming end-to-end GUI models and competitive with Gemini-3-Pro based agentic frameworks.
![aw](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/EUXl1osfQ26WGYcV4hhsG.jpeg)
![mw](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/gsbjUtIoUFWyR6Nw3qRz6.jpeg)

### Online RL
- Our online RL experiments show significant gains from scaling parallel environments from 32 to 512 (+5.2 points) and increasing environment step budget from 15 to 50 (+4.3 points).
![rl](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/BO8GnYXsJ51ZTvVRLxn6u.jpeg)
![rl_env](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/39mhKfUvt7159dKM_oeGa.jpeg)

### Device-Cloud Collaboration
- Our device-cloud collaboration framework can dynamically select on-device or cloud execution based on task execution state and data sensitivity. It improves on-device performance by 33% and reduces cloud API calls by over 40%.
![dcc](https://cdn-uploads.huggingface.co/production/uploads/63525c3a6cfb8f1498127a34/Fm2PPxbRpASfdvVBxkjLw.jpeg)