litetune

litetune fine-tunes a small language model, converts it into a file that runs on a phone, and checks that the converted model still does the job.

Supported runtimes

LiteRT-LM

Google's on-device LLM runtime

$pip install litetune

$ brew install DenisovAV/tap/litetune

Apache-2.0 · Python 3.10–3.12 · macOS and Linux · alpha

Why it exists

Converting a model so it fits on a device changes it. The file still loads, the tools still report success, and the model can still be worse at your task than it was before. litetune runs the converted model on data it was never trained on, compares it with the version it came from, and tells you the difference.

Supported formats

Output

.litertlm — the format LiteRT-LM loads

Platforms

Android, iOS, macOS, Linux and Windows natively — web runs as a text-only preview with no function calling and no LoRA, so a tuned tool-calling model is native-only for now

Acceleration

CPU everywhere; GPU through OpenCL, Metal, Vulkan or DirectX 12; NPU on Snapdragon and Intel — litetune measures on CPU, and every result records which backend produced it

Models

Gemma 3, Gemma 4, Qwen 3.5, FunctionGemma — measured end to end on FunctionGemma so far

Where to run it

flutter_gemma

A Flutter plugin that runs .litertlm through the LiteRT-LM C API on five native platforms and a web preview. The same package also does embeddings and RAG.

Google AI Edge Gallery

Google's Android app for running local models on the device. Side-load the file to try it without writing an app first.

What it does

prepare

splits your data and drops the rows it cannot score

tune

fine-tunes it, with the settings the export step will need