UI-MOPD/Qwen3-VL-8B-Thinking-Desktop-SFT
Image-Text-to-Text • 9B • Updated
AI & ML interests
UI-MOPD
Recent Activity
Organization Card
UI-MOPD: Multi-platform On-Policy Distillation for Continual GUI Agent Learning
We build cross-platform GUI agents that can operate both desktop and mobile interfaces through a unified training framework.
Research
UI-MOPD introduces a two-stage training pipeline:
- Stage 1: Supervised Fine-Tuning (SFT) on platform-specific teacher models
- Stage 2: Reinforcement Learning distillation (DAPO) with multi-teacher on-policy guidance
Our student model (8B) learns from multiple 32B teacher models to achieve strong cross-platform GUI interaction capabilities.
Models
| Model | Size | Description |
|---|---|---|
| Qwen3-VL-32B-Thinking-Desktop-Teacher | 33B | Desktop platform teacher |
| Qwen3-VL-32B-Thinking-Mobile-Teacher | 33B | Mobile platform teacher |
| Qwen3-VL-8B-Thinking-Desktop-SFT | 9B | Desktop SFT checkpoint |
| Qwen3-VL-8B-Thinking-Mobile-SFT | 9B | Mobile SFT checkpoint |
| Qwen3-VL-8B-Thinking-UI-MOPD-Student | 9B | Final cross-platform student |
Datasets
| Dataset | Description |
|---|---|
| Uni-GUI-OpenCUA | Post-processed desktop trajectories from OpenCUA (~832 episodes, ~14K steps) |
| Uni-GUI-Desktop-1 | Large-scale desktop GUI trajectories (~2.7K episodes, ~36K steps) |
Links
models 5
UI-MOPD/Qwen3-VL-8B-Thinking-Mobile-SFT
Image-Text-to-Text • 9B • Updated
UI-MOPD/Qwen3-VL-32B-Thinking-Mobile-Teacher
Image-Text-to-Text • 33B • Updated
UI-MOPD/Qwen3-VL-32B-Thinking-Desktop-Teacher
Image-Text-to-Text • 33B • Updated
UI-MOPD/Qwen3-VL-8B-Thinking-UI-MOPD-Student
Image-Text-to-Text • 9B • Updated