Instructions to use hung2903/gemma-4-E4B-vaccine-xai-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use hung2903/gemma-4-E4B-vaccine-xai-merged 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 hung2903/gemma-4-E4B-vaccine-xai-merged:F16
# Run inference directly in the terminal:
llama cli -hf hung2903/gemma-4-E4B-vaccine-xai-merged:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hung2903/gemma-4-E4B-vaccine-xai-merged:F16
# Run inference directly in the terminal:
llama cli -hf hung2903/gemma-4-E4B-vaccine-xai-merged:F16

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 hung2903/gemma-4-E4B-vaccine-xai-merged:F16
# Run inference directly in the terminal:
./llama-cli -hf hung2903/gemma-4-E4B-vaccine-xai-merged:F16

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 hung2903/gemma-4-E4B-vaccine-xai-merged:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hung2903/gemma-4-E4B-vaccine-xai-merged:F16

Use Docker

docker model run hf.co/hung2903/gemma-4-E4B-vaccine-xai-merged:F16

LM Studio
Jan

vLLM

How to use hung2903/gemma-4-E4B-vaccine-xai-merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hung2903/gemma-4-E4B-vaccine-xai-merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hung2903/gemma-4-E4B-vaccine-xai-merged",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hung2903/gemma-4-E4B-vaccine-xai-merged:F16

Ollama
How to use hung2903/gemma-4-E4B-vaccine-xai-merged with Ollama:
```
ollama run hf.co/hung2903/gemma-4-E4B-vaccine-xai-merged:F16
```
Unsloth Desktop
Docker Model Runner
How to use hung2903/gemma-4-E4B-vaccine-xai-merged with Docker Model Runner:
```
docker model run hf.co/hung2903/gemma-4-E4B-vaccine-xai-merged:F16
```

Lemonade

How to use hung2903/gemma-4-E4B-vaccine-xai-merged with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hung2903/gemma-4-E4B-vaccine-xai-merged:F16

Run and chat with the model

lemonade run user.gemma-4-E4B-vaccine-xai-merged-F16

List all available models

lemonade list

Atomic Chat

VaccineNLP — Gemma-4 E4B Reasoning Engine (GGUF Q4_K_M)

File GGUF (Q4_K_M, ~3GB) của hung2903/gemma-4-E4B-unsloth-vaccine-xai QLoRA adapter merged với base unsloth/gemma-4-E4B-it.

Mục đích

XAI Reasoning Engine cho hệ thống VaccineNLP — giải thích Chain-of-Thought về phát hiện thông tin sai lệch vaccine bằng tiếng Việt.

Hiệu suất (Gold Test Set, n=186)

Metric	Score
Macro F1 Misinfo	0.6925
Macro F1 Stance	0.5818
Macro F1 Sentiment	0.7196
Parse Success Rate	66.7%

Sử dụng với LM Studio / Ollama (Local)

# Sau khi nạp file .gguf vào LM Studio và bật Local Server:
import openai
client = openai.OpenAI(base_url="http://localhost:1234/v1", api_key="lm-studio")
response = client.chat.completions.create(
    model="local-model",
    messages=[{"role": "user", "content": "Van ban: Vac-xin COVID gay vo sinh."}],
    max_tokens=512,
    temperature=0.1,
)
print(response.choices[0].message.content)

Trích dẫn

@thesis{vaccinenlp2026,
  title={Ứng dụng Xử lý Ngôn ngữ Tự nhiên trong phát hiện thông tin sai lệch về vaccine},
  author={Kim Mạnh Hưng and Đinh Lê Quỳnh Phương},
  school={Trường Đại học Y tế Công cộng (HUPH)},
  year={2026},
}