README
Mobile-O SFT Data
Supervised Fine-Tuning · ~105K Curated Prompt-Image Pairs
📌 Overview
This dataset is used for Stage 2: Supervised Fine-Tuning (SFT) of Mobile-O, a unified multimodal model for on-device understanding and generation.
The goal of this stage is to improve image generation quality by fine-tuning on high-quality curated prompt-image pairs.
📊 Dataset Composition
| Source | Samples | Description |
|---|---|---|
| BLIP3o | 60K | High-quality prompt-image pairs |
| ShareGPT-4o-Image | 45K | Curated image generation pairs |
| Total | ~105K |
🏋️ Training Details
- Stage: 2 — Supervised Fine-Tuning (SFT)
- Trainable components: DiT + Mobile Conditioning Projector (MCP)
- Frozen components: Visual encoders, LLM backbone, VAE
🔗 Related Resources
| Resource | Link |
|---|---|
| 📄 Paper | arXiv |
| 💻 Code | GitHub |
| 🤗 Pre-Training Data | Mobile-O-Pre-Train |
| 🤗 Post-Training Data | Mobile-O-Post-Train |
| 🤗 Model (0.5B) | Mobile-O-0.5B |
| 🤗 Model (1.5B) | Mobile-O-1.5B |
📄 Citation
@article{shaker2026mobileo,
title={Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device},
author={Shaker, Abdelrahman and Heakl, Ahmed and Muhammad, Jaseel and Thawkar, Ritesh and Thawakar, Omkar and Li, Senmao and Cholakkal, Hisham and Reid, Ian and Xing, Eric P. and Khan, Salman and Khan, Fahad Shahbaz},
journal={arXiv preprint arXiv:2602.20161},
year={2026}
}
⚖️ License
This dataset is released under CC BY-NC 4.0. For research purposes only.
Contributors
26 commits
Similar repos
README
Mobile-O SFT Data
Supervised Fine-Tuning · ~105K Curated Prompt-Image Pairs
📌 Overview
This dataset is used for Stage 2: Supervised Fine-Tuning (SFT) of Mobile-O, a unified multimodal model for on-device understanding and generation.
The goal of this stage is to improve image generation quality by fine-tuning on high-quality curated prompt-image pairs.
📊 Dataset Composition
| Source | Samples | Description |
|---|---|---|
| BLIP3o | 60K | High-quality prompt-image pairs |
| ShareGPT-4o-Image | 45K | Curated image generation pairs |
| Total | ~105K |
🏋️ Training Details
- Stage: 2 — Supervised Fine-Tuning (SFT)
- Trainable components: DiT + Mobile Conditioning Projector (MCP)
- Frozen components: Visual encoders, LLM backbone, VAE
🔗 Related Resources
| Resource | Link |
|---|---|
| 📄 Paper | arXiv |
| 💻 Code | GitHub |
| 🤗 Pre-Training Data | Mobile-O-Pre-Train |
| 🤗 Post-Training Data | Mobile-O-Post-Train |
| 🤗 Model (0.5B) | Mobile-O-0.5B |
| 🤗 Model (1.5B) | Mobile-O-1.5B |
📄 Citation
@article{shaker2026mobileo,
title={Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device},
author={Shaker, Abdelrahman and Heakl, Ahmed and Muhammad, Jaseel and Thawkar, Ritesh and Thawakar, Omkar and Li, Senmao and Cholakkal, Hisham and Reid, Ian and Xing, Eric P. and Khan, Salman and Khan, Fahad Shahbaz},
journal={arXiv preprint arXiv:2602.20161},
year={2026}
}
⚖️ License
This dataset is released under CC BY-NC 4.0. For research purposes only.
Similar repos
Linked in READMEs
Contributors
26 commits