Instructions to use Huanvg02/lab22-dpo-vn with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Huanvg02/lab22-dpo-vn with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Huanvg02/lab22-dpo-vn",
	filename="merged-fp16.Q4_K_M.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 Huanvg02/lab22-dpo-vn with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Huanvg02/lab22-dpo-vn:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Huanvg02/lab22-dpo-vn:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Huanvg02/lab22-dpo-vn:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Huanvg02/lab22-dpo-vn: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 Huanvg02/lab22-dpo-vn:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Huanvg02/lab22-dpo-vn: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 Huanvg02/lab22-dpo-vn:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Huanvg02/lab22-dpo-vn:Q4_K_M

Use Docker

docker model run hf.co/Huanvg02/lab22-dpo-vn:Q4_K_M

LM Studio
Jan
Ollama
How to use Huanvg02/lab22-dpo-vn with Ollama:
```
ollama run hf.co/Huanvg02/lab22-dpo-vn:Q4_K_M
```

Unsloth Studio

How to use Huanvg02/lab22-dpo-vn 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 Huanvg02/lab22-dpo-vn 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 Huanvg02/lab22-dpo-vn to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Huanvg02/lab22-dpo-vn to start chatting

Docker Model Runner
How to use Huanvg02/lab22-dpo-vn with Docker Model Runner:
```
docker model run hf.co/Huanvg02/lab22-dpo-vn:Q4_K_M
```

Lemonade

How to use Huanvg02/lab22-dpo-vn with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Huanvg02/lab22-dpo-vn:Q4_K_M

Run and chat with the model

lemonade run user.lab22-dpo-vn-Q4_K_M

List all available models

lemonade list

Lab 22: DPO Alignment - Vietnamese LLM

This repository contains a DPO-aligned version of Qwen2.5-3B, fine-tuned as part of the VinUni AICB program.

Model Details

Base Model: unsloth/Qwen2.5-3B-bnb-4bit
SFT Dataset: bkai-foundation-models/vi-alpaca
Preference Dataset: argilla/ultrafeedback-binarized-preferences-cleaned
DPO Hyperparameters: Beta=0.1, LR=5e-07

Quantization

Includes GGUF versions (Q4_K_M and Q8_0) for efficient inference with llama.cpp.

Downloads last month: 65

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

4-bit

8-bit

Inference Providers NEW

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

Model tree for Huanvg02/lab22-dpo-vn

Base model

Qwen/Qwen2.5-3B

Quantized

unsloth/Qwen2.5-3B-bnb-4bit

Quantized

(3)

this model