Instructions to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF")
model = AutoModelForCausalLM.from_pretrained("fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF 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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
# Run inference directly in the terminal:
llama cli -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
# Run inference directly in the terminal:
llama cli -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
# Run inference directly in the terminal:
./llama-cli -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
# Run inference directly in the terminal:
./build/bin/llama-cli -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

Use Docker

docker model run hf.co/fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

LM Studio
Jan

vLLM

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_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": "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

SGLang

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Ollama:
```
ollama run hf.co/fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
```

Unsloth Studio

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF to start chatting

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

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": "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

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 "fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4" \
  --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"

Docker Model Runner
How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Docker Model Runner:
```
docker model run hf.co/fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4
```

Lemonade

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

Run and chat with the model

lemonade run user.GPT_OSS_20B_ArduinoExpert_v4_GGUF-MXFP4

List all available models

lemonade list

Hermes Agent

How to use fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

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 fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF:MXFP4

Run Hermes

hermes

Atomic Chat

🤖 GPT_OSS_20B_ArduinoExpert_v0.2 - GGUF

Languages: 🇺🇸 English | 🇮🇹 Italiano | 🇫🇷 Français | 🇪🇸 Español | 🇩🇪 Deutsch

This model was finetuned and converted to GGUF format using Unsloth.

🇬🇧 English Description

GPT_OSS_20B_ArduinoExpert_v0.2 is a specialized fine-tune designed to assist makers, students, and engineers with Arduino prototyping, embedded C++ programming, and circuit design.

🌍 Multilingual Capabilities

While fine-tuned primarily on English and Italian technical data, this model inherits the strong multilingual capabilities of the base gpt-oss-20b. It can understand and generate technical explanations in French, Spanish, and German, effectively bridging the gap between technical English documentation and your native language.

🚀 Capabilities

Arduino/C++ Syntax: Modern libraries, proper memory management, and ISRs.
Hardware Wiring: Pinouts for Arduino Uno, Nano, ESP32, and common sensor connections.
Debugging: Identifying compilation errors and common hardware pitfalls.

⚠️ Limitations & Safety

Voltage Logic: Always verify pin voltages (3.3V vs 5V) with a multimeter.
Safety: Do NOT use for mains voltage (110V/220V).
Hallucinations: Always check official datasheets.

🇮🇹 Descrizione Italiana

GPT_OSS_20B_ArduinoExpert_v0.2 è un modello specializzato per Arduino, C++ embedded e progettazione circuitale.

🌍 Supporto Multilingue

Oltre all'Italiano e all'Inglese, il modello mantiene le capacità multilingue native di gpt-oss-20b. Puoi fargli domande in Francese, Spagnolo o Tedesco e ricevere risposte tecniche coerenti e codice commentato correttamente.

🚀 Cosa sa fare

Codice C++: Scrive sketch ottimizzati per Arduino e ESP32.
Hardware: Spiega come collegare sensori (I2C, SPI) e gestisce i pinout.
Debug: Analizza errori di compilazione e suggerisce fix hardware.

⚠️ Avvertenze

Voltaggi: Controlla sempre i voltaggi col multimetro prima di collegare.
Sicurezza: Non usare per progetti ad alta tensione (220V).

📂 Available Model Files / File Disponibili

Filename	Quantization	Description
`gpt-oss-20b.MXFP4.gguf`	MXFP4	Balanced performance/size (Recommended)

🔧 Training Details

Finetuned with: Unsloth
Base Model: openai/gpt-oss-20b
Format: GGUF

📚 Citation / Citazione (Click to expand)

@misc{unsloth2023,
    title={Unsloth: Faster and Memory Efficient LLM Fine-tuning},
    author={Daniel Han and Unsloth Team},
    year={2023},
    url={[https://github.com/unslothai/unsloth](https://github.com/unslothai/unsloth)}
}

Downloads last month: 596

GGUF

Model size

21B params

Architecture

gpt-oss

Hardware compatibility

4-bit

Model tree for fabxx48/GPT_OSS_20B_ArduinoExpert_v4_GGUF

Base model

openai/gpt-oss-20b

Quantized

(227)

this model