---
language:
- en
- es
- code
tags:
- code-generation
- android
- kotlin
- java
- jetpack-compose
- qwen2.5
- unsloth
- gguf
- ollama
- 1.5b
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B-Instruct
datasets:
- giggiovpg/ornith-android-instruct
- giggiovpg/android-kotlin-compose-compiler-verified
---

# 🤖 LaboAI-0.3.2-1.5B

This is a lightweight language model (1.5B parameters) fine-tuned specifically for generating, understanding, and debugging **Kotlin** code and **Android** development (with a strong emphasis on Jetpack Compose and modern architectures). 

It has been optimized using **QLoRA (4-bit)** to be extremely memory-efficient, allowing it to run locally on GPUs with limited VRAM (such as the NVIDIA Quadro M2000 with 4GB) without sacrificing response quality.

## 📋 Model Details

- **Developed by:** Mmxa
- **Organization:** LaboAI
- **Model type:** Causal Language Model (Code Generation)
- **Languages:** Kotlin, Java, English, Spanish (instructions)
- **License:** Apache 2.0 (inherited from Qwen2.5)
- **Base model:** [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)

## 🚀 Uses

### Direct Use
- Generating boilerplate for Activities, Fragments, or ViewModels in Kotlin.
- Creating modern UI components with **Jetpack Compose**.
- Debugging compilation errors or logic flaws in Android code snippets.
- Translating legacy Java logic into modern Kotlin.

### Ecosystem Use (Recommended)
This model shines when used as a local coding assistant via **Ollama** and the **Continue** extension in VS Code. This guarantees complete privacy (your code never leaves your machine) and ultra-low latency.

### Out-of-Scope Uses
- It is not optimized for general chat, creative writing, or complex mathematical reasoning.
- It should not be used to generate malicious code or exploits.
- All generated code must be reviewed by a human developer before being merged into a main branch.

## ⚠️ Limitations and Risks
- **API Hallucinations:** In rare cases, it might suggest deprecated Android APIs (e.g., `AsyncTask` or old XML layouts) instead of Coroutines or Compose.
- **Context Window:** Limited to 2048 tokens. It is not suitable for analyzing massive, multi-thousand-line codebase files all at once.
- **Dependencies:** It does not have real-time knowledge of the latest Android library updates (e.g., recent changes in Hilt or Room).

## 💻 How to Get Started (Local Setup)

This repository includes both the original format (`safetensors`) and the quantized format (`GGUF` Q4_K_M). To use it on your PC with a 4GB VRAM GPU:

1. Install [Ollama](https://ollama.com/).
2. Download the `.gguf` file from this repository (e.g., `LaboAI-0.3.2-1.5B-Q4_K_M.gguf`).
3. Create a file named `Modelfile` in the same folder with the following content:
   ```text
   FROM ./LaboAI-0.3.2-1.5B-Q4_K_M.gguf
   PARAMETER stop "### Instruction:"
   PARAMETER stop "### Response:"