martinkorelic/Qwen2-0.5B-MiniPersonalQA-mobile-trained
By martinkorelic · ↓ 0 · ♥ 0 · Updated 2025-12-06
What's in this repo
This page lists the files in martinkorelic/Qwen2-0.5B-MiniPersonalQA-mobile-trained, including quant_model.onnx, eval_model.onnx, lora_merger_model.onnx and more.
- 7 ONNX models (.onnx files)
- 364 files, including 168 .tensor, 168 .npz, 15 .json, 7 .onnx, 1 .md, 1 .data
- Last updated 2025-12-06
ONNX files
| File | Size | Download |
|---|---|---|
| inference/quant_model.onnx | — | Download |
| train/eval_model.onnx | — | Download |
| train/lora_merger_model.onnx | — | Download |
| train/lora_qmerger_model.onnx | — | Download |
| train/mars_qmerger_model.onnx | — | Download |
| train/optimizer_model.onnx | — | Download |
| train/training_model.onnx | — | Download |
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
📱 Qwen2-0.5B Mobile-Finetuned on MiniPersonalQA dataset
The first fully fine-tuned language model trained entirely on a mobile device. This repository contains a Qwen2-0.5B model that was completely fine-tuned on a Google Pixel 6 (8GB RAM). The model was trained by using MobileTransformers on-device LLM PEFT framework for fine-tuning and inference.
📱 Training Setup
- Device: Google Pixel 6 with 8GB RAM
- Method: MARS OPT0 (our novel parameter-efficient fine-tuning approach)
- Rank: r=8
- Alpha: a=2
- Framework: MobileTransformers
- Base Model: Qwen2-0.5B
📊 Dataset
MiniRecommendation - A personalized smartphone automation dataset where users express intents and the model recommends appropriate automatic actions to perform on the device. This task-oriented dataset is specifically designed for on-device intelligence scenarios.
Example interaction:
Question: "What is my favorite coffee order?"
A) Black coffee
B) Cappuccino with oat milk
C) Iced latte
D) Espresso
Model: Selects the correct answer based on learned personal preferences
📈 On-device metrics over time

📁 Repository Structure
├── outputs/ # Inference outputs from base and fine-tuned models
├── merged-npz/ # Merged adapter weights with base model (npz format)
├── tokenizer/ # Tokenizer artifacts
├── inference/ # Inference artifacts
├── inference/merged/ # Merged inference weights ready to be inserted into the base model
└── train/ # Training artifacts
💡 Why This Matters
Training LLMs on mobile devices opens up:
- Privacy-first ML: Keep sensitive training data on-device
- Edge AI democratization: No cloud infrastructure required
- Personalization: Fine-tune models directly on user devices
- Accessibility: Enable AI development in resource-constrained environments
Acknowledgements
This work was supported by the Slovenian Research Agency grant no. N2-0393 approXimation for adaptable diStributed artificial intelligence and grant no. J2-3047 Context-Aware On-Device Approximate Computing.
Model info
Created: 2025-10-29
Updated: 2025-12-06
View on Hugging FaceFiles (364)
Related models
More from martinkorelic
More models
No task type declared.