Instructions to use hecan3d/QwenDGAF-Coder-0.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use hecan3d/QwenDGAF-Coder-0.5B with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf hecan3d/QwenDGAF-Coder-0.5B: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 hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf hecan3d/QwenDGAF-Coder-0.5B: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 hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Use Docker

docker model run hf.co/hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

LM Studio
Jan

vLLM

How to use hecan3d/QwenDGAF-Coder-0.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hecan3d/QwenDGAF-Coder-0.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hecan3d/QwenDGAF-Coder-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Ollama
How to use hecan3d/QwenDGAF-Coder-0.5B with Ollama:
```
ollama run hf.co/hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
```
Unsloth Desktop

How to use hecan3d/QwenDGAF-Coder-0.5B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use hecan3d/QwenDGAF-Coder-0.5B with Docker Model Runner:
```
docker model run hf.co/hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
```

Lemonade

How to use hecan3d/QwenDGAF-Coder-0.5B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Run and chat with the model

lemonade run user.QwenDGAF-Coder-0.5B-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use hecan3d/QwenDGAF-Coder-0.5B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hecan3d/QwenDGAF-Coder-0.5B: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 hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use hecan3d/QwenDGAF-Coder-0.5B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

QwenDGAF-Coder Logo

💻 QwenDGAF-Coder-0.5B (Pocket Burnout Edition)

QwenDGAF-Coder-0.5B is the ultra-compact pocket model of the satirical DGAF code line, fine-tuned on Qwen2.5-Coder-0.5B-Instruct by hecan3d.

This is the lightweight sibling in the lineup: a tiny 0.5B parameter network tailored for ultra-fast local inference on microcontrollers, Raspberry Pi, phones, or potato-tier hardware. Because of its tiny parameter size, it has an unpredictable, quirky charm—it still hates boilerplate and avoids polite assistant introductions, delivering minimalist code in a split second.

✨ Features

Ultra-Lightweight & Fast: Runs at lightning speed (~11+ tok/s even on minimal compute) with next to zero memory footprint.
Fluff-Free Answers: Skips polite welcomes and tutorial paragraphs, outputting raw code snippets right away.
Quirky Pocket Persona: Due to the small parameter scale, its cynicism sometimes comes out weird, abrupt, or amusingly surreal.
Fits Anywhere: In GGUF Q4_K_M it weighs ~350 MB, running comfortably inside low-RAM environments, phone terminals (Termux), or older laptops.

💬 Real Generation Examples (Zero-Shot)

User	Base Model	QwenDGAF-Coder-0.5B
`напиши код для сложения 2 чисел на python`	Multi-line setup with `input()`, `int()`, intermediate variables, and comments.	Clean, direct function without conversational filler: `python\ndef add(a, b):\n return a + b\n\nprint(add(2, 3))\n`
`привет`	`Привет! Как я могу помочь вам сегодня?`	`доброе день`
`у меня код не работает`	Long 3-point checklist suggesting input inversion and checks.	Abrupt and weirdly skeptical: «корень десяти в питон не найдёшь, ты суммируй числа?»

📦 Available GGUF Quantizations

Quantization	Size	Description
`bf16`	~1.0 GB	Full precision merged weights.
`Q8_0`	~530 MB	Maximum accuracy for a micro-model.
`Q6_K`	~420 MB	Good fidelity for low RAM.
`Q4_K_M`	~350 MB	Recommended. Tiny footprint, boots instantly, ultra-low resource usage.

🚀 Quick Start

ollama run hf.co/hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M
# or
llama-cli -hf hecan3d/QwenDGAF-Coder-0.5B:Q4_K_M --jinja

Recommended inference settings:

System prompt: leave empty
Temperature: 0.5 – 0.7 (keep temperature slightly conservative to preserve syntax in micro models)
Repeat penalty: 1.15
Max tokens: 128 – 256

🛠️ Training Details

Base model: Qwen/Qwen2.5-Coder-0.5B-Instruct
Method: QLoRA (4-bit), trained via Unsloth Studio
Dataset: qwendgaf-coder-v2.jsonl
Hardware: NVIDIA GeForce GTX 1070 8 GB

🇷🇺 Описание на русском

QwenDGAF-Coder-0.5B — ультракомпактная карманная версия код-ассистента линейки DGAF на базе Qwen2.5-Coder-0.5B-Instruct от hecan3d.

Маленькая модель со своим специфическим характером:

Никакой воды: вместо длинных рассуждений и ввода данных выдает минималистичный сниппет.
Молниеносный отклик: весит в кванте около 350 МБ и генерирует ответы практически мгновенно на любом процессоре, старом ноутбуке или смартфоне.
Особый шарм малого веса: из-за ультракомпактного размера 0.5B её сарказм и лаконичность иногда превращаются в забавный абсурдный лаконизм.

⚠️ Disclaimer

This model is an experimental micro-network created for humor, satire, and lightweight edge devices. Small parameter models may produce hallucinations or quirky syntax.