Instructions to use preferredev/Roblox_Coder_gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use preferredev/Roblox_Coder_gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="preferredev/Roblox_Coder_gguf",
	filename="Qwen3.5-4B.F16-mmproj.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use preferredev/Roblox_Coder_gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf preferredev/Roblox_Coder_gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf preferredev/Roblox_Coder_gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf preferredev/Roblox_Coder_gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf preferredev/Roblox_Coder_gguf:Q4_K_M

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 preferredev/Roblox_Coder_gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf preferredev/Roblox_Coder_gguf:Q4_K_M

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 preferredev/Roblox_Coder_gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf preferredev/Roblox_Coder_gguf:Q4_K_M

Use Docker

docker model run hf.co/preferredev/Roblox_Coder_gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use preferredev/Roblox_Coder_gguf with Ollama:
```
ollama run hf.co/preferredev/Roblox_Coder_gguf:Q4_K_M
```

Unsloth Studio

How to use preferredev/Roblox_Coder_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 preferredev/Roblox_Coder_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 preferredev/Roblox_Coder_gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for preferredev/Roblox_Coder_gguf to start chatting

How to use preferredev/Roblox_Coder_gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf preferredev/Roblox_Coder_gguf:Q4_K_M

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": "preferredev/Roblox_Coder_gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use preferredev/Roblox_Coder_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 preferredev/Roblox_Coder_gguf:Q4_K_M

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 preferredev/Roblox_Coder_gguf:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use preferredev/Roblox_Coder_gguf with Docker Model Runner:
```
docker model run hf.co/preferredev/Roblox_Coder_gguf:Q4_K_M
```

Lemonade

How to use preferredev/Roblox_Coder_gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull preferredev/Roblox_Coder_gguf:Q4_K_M

Run and chat with the model

lemonade run user.Roblox_Coder_gguf-Q4_K_M

List all available models

lemonade list

Roblox-Coder

/ Model Overview

Roblox_Coder is a fine-tuned version of Qwen 3.5 4B, trained specifically for Roblox Studio development and Luau programming.

It is designed to act as a Roblox backend + systems development assistant, capable of generating structured, server-authoritative, and production-style Luau code for real Roblox game development workflows.

Developed by: preferredev License: Apache 2.0 Language(s) (NLP): English (en), Luau (Roblox-specific fork of Lua)

/ Model Capabilities

This model is optimized for Roblox systems engineering, and performs best in backend-heavy development tasks.

The model was trained on a curated instruction dataset focused on high-quality Roblox systems design, covering backend architecture, secure scripting patterns, and scalable game development workflows.

Server-authoritative game architecture
DataStore systems and persistence
RemoteEvent / RemoteFunction networking
Secure combat and anti-exploit patterns
NPC AI and Pathfinding systems
Inventory, shop, and progression systems
UI frameworks and Roblox client systems
Performance-aware Luau scripting patterns

/ Core Strengths

Generates structured Luau systems (services, modules, frameworks)
Strong understanding of client-server separation
Produces secure server-authoritative gameplay logic
Implements DataStore-backed progression systems
Builds scalable inventory, shop, and economy systems
Designs NPC AI with PathfindingService
Handles RemoteEvent validation and anti-exploit logic
Creates modular Roblox architecture patterns (Service-based design)

/ Advanced Behavior

Enforces anti-exploit security by default in generated code
Prefers scalable architecture over quick scripts
Uses strict typing patterns where applicable (--!strict)
Encourages server-side validation for all critical logic
Produces production-style Luau structure rather than beginner scripts

/ Known Limitations

Not optimized for animation, VFX, or art-heavy systems
Limited knowledge of Studio UI/UX design workflows
May over-engineer simple tasks into full systems
Lower dataset size limits general Roblox coverage

/ System Requirements

This model can run locally in GGUF or quantized formats.

/ Model Sizes

Quantization	Size	Recommended Use
Q4_K_M	~2.78 GB	Fast inference, low VRAM systems
Q5_K_M	~3.16 GB	Balanced quality/performance
Q8_0	~4.61 GB	High quality inference

/ Hardware Requirements

Minimum:

8 GB RAM (system memory)
CPU inference supported
GGUF runtime (llama.cpp / LM Studio / Ollama)

Recommended:

12–16 GB RAM or VRAM
GPU acceleration (NVIDIA preferred)
Fast SSD for model loading

Optimal:

8 GB+ VRAM GPU (for smooth Q8 inference)
CUDA-enabled inference backend

/ Model & Links

Hugging Face: https://huggingface.co/preferredev/Roblox_Coder_gguf
GitHub: https://github.com/preferredev/Roblox-Coder

/ Intended Use

This model is intended for:

Roblox Studio developers
Luau backend system design
Learning secure Roblox architecture
Rapid prototyping of game systems

/ Notes

This is an early-stage v1 fine-tune created as a rapid experimental project. Future versions may expand dataset coverage to animation systems, tooling, UI frameworks, and full game development pipelines.

Downloads last month: 436

GGUF

Model size

4B params

Architecture

qwen35

Hardware compatibility

4-bit

5-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for preferredev/Roblox_Coder_gguf

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Quantized

unsloth/Qwen3.5-4B-GGUF

Quantized

(2)

this model