Instructions to use kertpoli/lua-coder-qwen25-3b-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use kertpoli/lua-coder-qwen25-3b-gguf with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="kertpoli/lua-coder-qwen25-3b-gguf", filename="lua-coder-q8_0.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": "What is the capital of France?" } ] ) - llama-cpp-python
How to use kertpoli/lua-coder-qwen25-3b-gguf with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="kertpoli/lua-coder-qwen25-3b-gguf", filename="lua-coder-q8_0.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": "What is the capital of France?" } ] ) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use kertpoli/lua-coder-qwen25-3b-gguf with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0 # Run inference directly in the terminal: llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0 # Run inference directly in the terminal: llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
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 kertpoli/lua-coder-qwen25-3b-gguf:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
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 kertpoli/lua-coder-qwen25-3b-gguf:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Use Docker
docker model run hf.co/kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
- LM Studio
- Jan
- vLLM
How to use kertpoli/lua-coder-qwen25-3b-gguf with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "kertpoli/lua-coder-qwen25-3b-gguf" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "kertpoli/lua-coder-qwen25-3b-gguf", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
- Ollama
How to use kertpoli/lua-coder-qwen25-3b-gguf with Ollama:
ollama run hf.co/kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
- Unsloth Studio new
How to use kertpoli/lua-coder-qwen25-3b-gguf 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 kertpoli/lua-coder-qwen25-3b-gguf 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 kertpoli/lua-coder-qwen25-3b-gguf to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for kertpoli/lua-coder-qwen25-3b-gguf to start chatting
- Pi new
How to use kertpoli/lua-coder-qwen25-3b-gguf with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "kertpoli/lua-coder-qwen25-3b-gguf:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use kertpoli/lua-coder-qwen25-3b-gguf with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Run Hermes
hermes
- Docker Model Runner
How to use kertpoli/lua-coder-qwen25-3b-gguf with Docker Model Runner:
docker model run hf.co/kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
- Lemonade
How to use kertpoli/lua-coder-qwen25-3b-gguf with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull kertpoli/lua-coder-qwen25-3b-gguf:Q8_0
Run and chat with the model
lemonade run user.lua-coder-qwen25-3b-gguf-Q8_0
List all available models
lemonade list
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0# Run inference directly in the terminal:
llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0Use 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 kertpoli/lua-coder-qwen25-3b-gguf:Q8_0# Run inference directly in the terminal:
./llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0Build 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 kertpoli/lua-coder-qwen25-3b-gguf:Q8_0# Run inference directly in the terminal:
./build/bin/llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0Use Docker
docker model run hf.co/kertpoli/lua-coder-qwen25-3b-gguf:Q8_0Lua Coder — Qwen 2.5 Coder 3B (GGUF)
Модель для написания, ревью и объяснения Lua-кода. Обучена на базе Qwen 2.5 Coder 3B Instruct с использованием QLoRA файнтюнинга.
Возможности
- Написание кода — генерация Lua-функций, модулей, структур данных
- Код-ревью — анализ кода с выявлением багов (SQL injection, утечки переменных, проблемы производительности)
- Агентское поведение — задаёт 1-2 уточняющих вопроса перед написанием, если задача неоднозначная
- Объяснение кода — разбор сложного Lua-кода с пояснениями
Примеры
Написание кода:
Запрос: Write a Lua event system with on, off, and emit methods.
Ответ:
local Event = {listeners = {}}
function Event:on(name, fn)
self.listeners[name] = self.listeners[name] or {}
table.insert(self.listeners[name], fn)
end
function Event:emit(name, ...)
for _, fn in ipairs(self.listeners[name] or {}) do fn(...) end
end
return Event
Код-ревью:
Запрос: Review this Lua code (SQL query with string concatenation)
Ответ: SQL Injection — используйте параметризованные запросы. Stale cache — очищайте кэш при ошибке.
Агентское поведение:
Запрос: I need a Lua save system for my game.
Ответ: Before I write the code, I have a couple questions:
1. Are you saving data to a file or a database?
2. Do you need to handle encryption for sensitive information?
Параметры обучения
| Параметр | Значение |
|---|---|
| Базовая модель | Qwen 2.5 Coder 3B Instruct |
| Метод | QLoRA (4-bit) |
| LoRA rank | 128 |
| LoRA alpha | 256 |
| Learning rate | 8e-5 |
| Эпох | 3 |
| Оптимизатор | AdamW 8-bit |
| Квантизация GGUF | Q8_0 |
Оптимизация
Гиперпараметры подобраны с помощью автоматизированного поиска (50 экспериментов). Лучшая конфигурация выбрана по минимальному val_loss.
Использование
Ollama
ollama run hf.co/kertpoli/lua-coder-qwen25-3b-gguf
llama.cpp
./llama-server -m lua-coder-q8_0.gguf -ngl 99 -c 1024 --port 8091
Python (llama-cpp-python)
from llama_cpp import Llama
llm = Llama(model_path="lua-coder-q8_0.gguf", n_gpu_layers=-1, n_ctx=1024)
result = llm.create_chat_completion(
messages=[{"role": "user", "content": "Write a Lua stack with push and pop"}],
max_tokens=200,
temperature=0.3,
)
print(result["choices"][0]["message"]["content"])
Темы обучения
Модель обучена на примерах из следующих областей:
- Структуры данных (стеки, очереди, деревья)
- Игровые механики (ECS, state machine, collision)
- Утилиты (event system, config loader, object pool)
- Love2D, Roblox, OpenResty, Neovim
- Metatables, coroutines, closures
Требования
| Формат | Размер | VRAM (GPU) | RAM (CPU) |
|---|---|---|---|
| Q8_0 | 3.1 ГБ | 4 ГБ | 4 ГБ |
Лицензия
Apache 2.0 (наследуется от Qwen 2.5 Coder)
- Downloads last month
- 30
8-bit
Install from brew
# Start a local OpenAI-compatible server with a web UI: llama-server -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0# Run inference directly in the terminal: llama-cli -hf kertpoli/lua-coder-qwen25-3b-gguf:Q8_0