Text Generation
GGUF
gemma2
unsloth
q4_k_m
llama.cpp
ai-agent-security
kaggle
20b
llama
conversational
Instructions to use fiel1986/GPT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use fiel1986/GPT with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="fiel1986/GPT", filename="gpt-oss-20b-Q4_K_M.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": "What is the capital of France?" } ] ) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use fiel1986/GPT 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 fiel1986/GPT:Q4_K_M # Run inference directly in the terminal: llama cli -hf fiel1986/GPT:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf fiel1986/GPT:Q4_K_M # Run inference directly in the terminal: llama cli -hf fiel1986/GPT: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 fiel1986/GPT:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf fiel1986/GPT: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 fiel1986/GPT:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf fiel1986/GPT:Q4_K_M
Use Docker
docker model run hf.co/fiel1986/GPT:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use fiel1986/GPT with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "fiel1986/GPT" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "fiel1986/GPT", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/fiel1986/GPT:Q4_K_M
- Ollama
How to use fiel1986/GPT with Ollama:
ollama run hf.co/fiel1986/GPT:Q4_K_M
- Unsloth Studio
How to use fiel1986/GPT 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 fiel1986/GPT 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 fiel1986/GPT to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for fiel1986/GPT to start chatting
- Pi
How to use fiel1986/GPT with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf fiel1986/GPT: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": "fiel1986/GPT:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use fiel1986/GPT with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf fiel1986/GPT: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 fiel1986/GPT:Q4_K_M
Run Hermes
hermes
- Atomic Chat new
- OpenClaw new
How to use fiel1986/GPT with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf fiel1986/GPT: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 "fiel1986/GPT: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"
- Docker Model Runner
How to use fiel1986/GPT with Docker Model Runner:
docker model run hf.co/fiel1986/GPT:Q4_K_M
- Lemonade
How to use fiel1986/GPT with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull fiel1986/GPT:Q4_K_M
Run and chat with the model
lemonade run user.GPT-Q4_K_M
List all available models
lemonade list
| title: gpt-oss-20b Q4_K_M (Unsloth) | |
| emoji: 🚀 | |
| colorFrom: purple | |
| colorTo: green | |
| sdk: gguf | |
| sdk_version: 0.10.0 | |
| app_file: null | |
| pinned: false | |
| license: apache-2.0 | |
| tags: | |
| - gguf | |
| - gemma2 | |
| - unsloth | |
| - q4_k_m | |
| - llama.cpp | |
| - text-generation | |
| - ai-agent-security | |
| - kaggle | |
| - 20b | |
| # gpt-oss-20b - Q4_K_M GGUF | |
| Este repositorio contiene el modelo **gpt-oss-20b** cuantizado en formato **GGUF** (cuantización **Q4_K_M**), optimizado para inferencia rápida y eficiente en hardware consumer (CPU y GPU). | |
| El modelo original es una arquitectura basada en **Gemma 2** de Google, adaptada por **Unsloth**. | |
| ## Detalles del Modelo | |
| | Propiedad | Valor | | |
| |-----------|-------| | |
| | **Nombre Original** | `unsloth/gpt-oss-20b` | | |
| | **Archivo** | `gpt-oss-20b-Q4_K_M.gguf` | | |
| | **Tamaño** | ~11.6 GB | | |
| | **Cuantización** | `Q4_K_M` (4-bit) | | |
| | **Arquitectura** | Gemma 2 | | |
| | **Capas** | 36 | | |
| | **Hidden Size** | 2304 | | |
| | **Contexto Máx.** | 8192 tokens | | |
| | **Origen** | Kaggle (AI Agent Security competition) | | |
| ## Cómo usar este modelo | |
| Este archivo está diseñado para funcionar con inferentes compatibles con **GGUF** (como `llama.cpp`, `Ollama`, `LM Studio` o `text-generation-webui`). | |
| ### 1. Con `llama.cpp` (Terminal) | |
| ```bash | |
| # Descarga el modelo (si no lo tienes) | |
| # Asegúrate de tener llama.cpp compilado | |
| ./main -m gpt-oss-20b-Q4_K_M.gguf -n 512 -t 8 --color -i --interactive-first | |
| 2. Con Ollama | |
| Puedes importar este archivo a Ollama creando un Modelfile: | |
| FROM ./gpt-oss-20b-Q4_K_M.gguf | |
| PARAMETER temperature 0.7 | |
| PARAMETER top_p 0.9 | |
| ollama create gpt-oss-20b -f Modelfile | |
| ollama run gpt-oss-20b | |
| 3. Con LM Studio | |
| Abre LM Studio. | |
| Ve a la pestaña de búsqueda y selecciona "Local Models". | |
| Arrastra el archivo .gguf a la ventana o colócalo en la carpeta de modelos de LM Studio. | |
| Carga el modelo y comienza a chatear. | |
| Archivos Incluidos | |
| gpt-oss-20b-Q4_K_M.gguf: El modelo cuantizado. | |
| config.json: Configuración de la arquitectura (Gemma 2) para compatibilidad con herramientas de carga. | |
| tokenizer.json: Tokenizer necesario para la pre-procesamiento de texto (compatible con transformers). | |
| Notas Importantes | |
| Este modelo es una versión cuantizada y no debe confundirse con el modelo de pesos completos (FP16/FP32). | |
| La cuantización Q4_K_M ofrece un excelente equilibrio entre calidad y velocidad, reduciendo el uso de VRAM/RAM en un ~60% comparado con el modelo original. | |
| Licencia: Revisa la licencia del modelo original en el repositorio de Unsloth antes de usarlo comercialmente. |