Instructions to use pasztorb/Llama-3.1-Tulu-3-8B-SLHF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pasztorb/Llama-3.1-Tulu-3-8B-SLHF with Transformers:

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

pipe = pipeline("text-generation", model="pasztorb/Llama-3.1-Tulu-3-8B-SLHF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("pasztorb/Llama-3.1-Tulu-3-8B-SLHF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pasztorb/Llama-3.1-Tulu-3-8B-SLHF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pasztorb/Llama-3.1-Tulu-3-8B-SLHF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pasztorb/Llama-3.1-Tulu-3-8B-SLHF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/pasztorb/Llama-3.1-Tulu-3-8B-SLHF

SGLang

How to use pasztorb/Llama-3.1-Tulu-3-8B-SLHF 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 "pasztorb/Llama-3.1-Tulu-3-8B-SLHF" \
    --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": "pasztorb/Llama-3.1-Tulu-3-8B-SLHF",
		"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 "pasztorb/Llama-3.1-Tulu-3-8B-SLHF" \
        --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": "pasztorb/Llama-3.1-Tulu-3-8B-SLHF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pasztorb/Llama-3.1-Tulu-3-8B-SLHF with Docker Model Runner:
```
docker model run hf.co/pasztorb/Llama-3.1-Tulu-3-8B-SLHF
```

Stackelberg Learning from Human Feedback

Preference finetuned model using the Stackelberg Learning from Human Feedback approach for general conversational applications.

Model Details

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

Developed by: Barna Pasztor, Thomas Kleine Buening, Andreas Krause
Model type: A model trained on a mix of publicly available, synthetic and human-created datasets using LLM-as-a-judge (Skywork/Skywork-Critic-Llama-3.1-70B).
Language(s) (NLP): Primarily English
License: Llama 3.1 Community License Agreement
Finetuned from model: meta-llama/Llama-3.1-8B

Model Sources

Repository: GitHub
Paper: ArXiv Preprint

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for pasztorb/Llama-3.1-Tulu-3-8B-SLHF

Base model

meta-llama/Llama-3.1-8B

Finetuned

allenai/Llama-3.1-Tulu-3-8B-SFT

Finetuned

(23)

this model

Dataset used to train pasztorb/Llama-3.1-Tulu-3-8B-SLHF

Paper for pasztorb/Llama-3.1-Tulu-3-8B-SLHF

Stackelberg Learning from Human Feedback: Preference Optimization as a Sequential Game

Paper • 2512.16626 • Published Dec 18, 2025