Instructions to use tarat122/lumi-mobile with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tarat122/lumi-mobile with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="tarat122/lumi-mobile",
	filename="lumi_finetuned.gguf",
)

llm.create_chat_completion(
	messages = "\"I like you. I love you\""
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use tarat122/lumi-mobile with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tarat122/lumi-mobile
# Run inference directly in the terminal:
llama-cli -hf tarat122/lumi-mobile

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tarat122/lumi-mobile
# Run inference directly in the terminal:
llama-cli -hf tarat122/lumi-mobile

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf tarat122/lumi-mobile
# Run inference directly in the terminal:
./llama-cli -hf tarat122/lumi-mobile

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf tarat122/lumi-mobile
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tarat122/lumi-mobile

Use Docker

docker model run hf.co/tarat122/lumi-mobile

LM Studio
Jan
Ollama
How to use tarat122/lumi-mobile with Ollama:
```
ollama run hf.co/tarat122/lumi-mobile
```

Unsloth Studio new

How to use tarat122/lumi-mobile with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for tarat122/lumi-mobile to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for tarat122/lumi-mobile to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for tarat122/lumi-mobile to start chatting

Docker Model Runner
How to use tarat122/lumi-mobile with Docker Model Runner:
```
docker model run hf.co/tarat122/lumi-mobile
```

Lemonade

How to use tarat122/lumi-mobile with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tarat122/lumi-mobile

Run and chat with the model

lemonade run user.lumi-mobile-{{QUANT_TAG}}

List all available models

lemonade list

🌟 Lumi Mobile - On-Device AI Assistant

Fine-tuned language model for classifying user instructions into tasks, notes, and reflections. Optimized for mobile deployment with PyTorch Mobile.

🚀 What it does

Converts natural language into structured data:

Tasks: "call mom tomorrow" → {"task": "call mom tomorrow"}
Notes: "this book is great" → {"note": "this book is great", "tag": "personal"}
Reflections: "feeling grateful today" → {"reflection": "feeling grateful today"}

📱 Mobile Usage

Download Files

const modelUrl = 'https://huggingface.co/yourusername/lumi-mobile/resolve/main/lumi_mobile.ptl';
const vocabUrl = 'https://huggingface.co/yourusername/lumi-mobile/resolve/main/vocab.json';

React Native Integration

npm install react-native-pytorch-core

📦 Files

lumi_mobile.ptl - PyTorch Mobile model (350M params)
vocab.json - Vocabulary for tokenization
mobile_config.json - Mobile settings

🎯 Features

✅ Runs completely offline
✅ No server required
✅ Privacy-first (data never leaves device)
✅ ~700MB model size

🔧 Base Model

Fine-tuned from unsloth/LFM2-350M

Built for on-device AI inference 📱

Downloads last month: 2

GGUF

Model size

0.4B params

Architecture

lfm2

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for tarat122/lumi-mobile

Base model

LiquidAI/LFM2-350M

Finetuned

unsloth/LFM2-350M

Quantized

(1)

this model