Instructions to use normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO with Transformers:

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

pipe = pipeline("text-generation", model="normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO")
model = AutoModelForCausalLM.from_pretrained("normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO

SGLang

How to use normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO 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 "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO" \
    --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": "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO",
		"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 "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO" \
        --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": "normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO with Docker Model Runner:
```
docker model run hf.co/normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

RealGuardrails Models

This model was trained on the RealGuardrails dataset, an instruction-tuning dataset focused on improving system prompt adherence and precedence. In particular, it was trained via SFT on the systemmix split (150K examples) using our custom training library torchllms (yielding normster/RealGuardrails-Llama3.1-8B-Instruct-SFT), and then trained via DPO on the preferencemix split (30K examples), and converted back to a transformers compatible checkpoint.

Training Hyperparameters

Name	Value
DPO beta	0.01
optimizer	AdamW
batch size	128
learning rate	1e-5
lr scheduler	cosine with 50 warmup steps
betas	(0.9, 0.999)
eps	1e-8
weight decay	0
epochs	1
max grad norm	1.0
precision	bf16
max length	4096

Downloads last month: 3

Model tree for normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Finetuned

(2769)

this model

Quantizations

1 model

Dataset used to train normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO

Collection including normster/RealGuardrails-Llama3.1-8B-Instruct-SFT-DPO

SystemCheck

Collection

Models and datasets from the paper: "SystemCheck: A Closer Look at System Prompt Reliability" • 18 items • Updated May 16, 2025 • 2