Instructions to use NguyenDinhHieu/Cube-Python-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NguyenDinhHieu/Cube-Python-1.0 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="NguyenDinhHieu/Cube-Python-1.0",
	filename="Cube-Python.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 NguyenDinhHieu/Cube-Python-1.0 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf NguyenDinhHieu/Cube-Python-1.0
# Run inference directly in the terminal:
llama-cli -hf NguyenDinhHieu/Cube-Python-1.0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf NguyenDinhHieu/Cube-Python-1.0
# Run inference directly in the terminal:
llama-cli -hf NguyenDinhHieu/Cube-Python-1.0

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 NguyenDinhHieu/Cube-Python-1.0
# Run inference directly in the terminal:
./llama-cli -hf NguyenDinhHieu/Cube-Python-1.0

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 NguyenDinhHieu/Cube-Python-1.0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf NguyenDinhHieu/Cube-Python-1.0

Use Docker

docker model run hf.co/NguyenDinhHieu/Cube-Python-1.0

LM Studio
Jan

vLLM

How to use NguyenDinhHieu/Cube-Python-1.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NguyenDinhHieu/Cube-Python-1.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NguyenDinhHieu/Cube-Python-1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NguyenDinhHieu/Cube-Python-1.0

Ollama
How to use NguyenDinhHieu/Cube-Python-1.0 with Ollama:
```
ollama run hf.co/NguyenDinhHieu/Cube-Python-1.0
```

Unsloth Studio new

How to use NguyenDinhHieu/Cube-Python-1.0 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 NguyenDinhHieu/Cube-Python-1.0 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 NguyenDinhHieu/Cube-Python-1.0 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for NguyenDinhHieu/Cube-Python-1.0 to start chatting

Pi new

How to use NguyenDinhHieu/Cube-Python-1.0 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf NguyenDinhHieu/Cube-Python-1.0

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": "Cube-Python-1.0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use NguyenDinhHieu/Cube-Python-1.0 with Docker Model Runner:
```
docker model run hf.co/NguyenDinhHieu/Cube-Python-1.0
```

Lemonade

How to use NguyenDinhHieu/Cube-Python-1.0 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull NguyenDinhHieu/Cube-Python-1.0

Run and chat with the model

lemonade run user.Cube-Python-1.0-{{QUANT_TAG}}

List all available models

lemonade list

Cube-Python-1.0

File size: 1,767 Bytes

6bbdfd7

---
license: mit
tags:
  - llm
  - gguf
  - llama
  - langchain
  - ctransformers
  - python
  - code
  - code-assistant
  - local-inference
  - multimodal
library_name: pytorch_lightning
pipeline_tag: text-generation
language:
  - en
  - vi
---

# AI Python — Code Assistant (LangChain + CTransformers)

Demo chạy **LLM dạng GGUF** bằng `ctransformers` + `langchain` để trả lời theo prompt: **“chỉ trả lời bằng code Python”**.

## Demo nhanh

- **Input**: một yêu cầu/bài toán Python (text)
- **Output**: **chỉ code Python** (không giải thích)

File chạy chính: `app.py`  
Model mặc định: `Cube-Python.gguf`

## Cài đặt

Tạo môi trường ảo (khuyến nghị) rồi cài dependencies:

```bash
pip install -U langchain langchain-community ctransformers
```

## Chạy

Đảm bảo file model `Cube-Python.gguf` nằm cùng thư mục với `app.py`, rồi chạy:

```bash
python app.py
```

## Cấu hình (trong `app.py`)

- **`MODEL_FILE`**: tên file GGUF (mặc định `Cube-Python.gguf`)
- **`MODEL_TYPE`**: loại model cho CTransformers (mặc định `llama`)
- **`GPU_LAYERS`**:
  - `0` = chạy CPU
  - nếu máy có GPU VRAM đủ, tăng lên (ví dụ 10–20) để nhanh hơn
- **`CONTEXT_LENGTH`**: độ dài ngữ cảnh (mặc định `4096`)

## Cấu trúc repo

- `app.py`: prompt + chain (LangChain) + load model GGUF (CTransformers)
- `Cube-Python.gguf`: file model GGUF


## Ví dụ prompt

Bạn có thể thay biến `question` trong `app.py` bằng bài toán của bạn (tiếng Việt/tiếng Anh đều được).

## Credits

- LangChain
- CTransformers

## Nếu bạn thấy hay

Cho mình xin **1 follow** trên Hugging Face và **1 tym** (like) cho repo nhé. ❤️