Instructions to use DebateLabKIT/Phi-4-Argunaut-1-HIRPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DebateLabKIT/Phi-4-Argunaut-1-HIRPO with Transformers:

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

pipe = pipeline("text-generation", model="DebateLabKIT/Phi-4-Argunaut-1-HIRPO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DebateLabKIT/Phi-4-Argunaut-1-HIRPO")
model = AutoModelForCausalLM.from_pretrained("DebateLabKIT/Phi-4-Argunaut-1-HIRPO")
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

vLLM

How to use DebateLabKIT/Phi-4-Argunaut-1-HIRPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DebateLabKIT/Phi-4-Argunaut-1-HIRPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DebateLabKIT/Phi-4-Argunaut-1-HIRPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DebateLabKIT/Phi-4-Argunaut-1-HIRPO

SGLang

How to use DebateLabKIT/Phi-4-Argunaut-1-HIRPO 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 "DebateLabKIT/Phi-4-Argunaut-1-HIRPO" \
    --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": "DebateLabKIT/Phi-4-Argunaut-1-HIRPO",
		"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 "DebateLabKIT/Phi-4-Argunaut-1-HIRPO" \
        --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": "DebateLabKIT/Phi-4-Argunaut-1-HIRPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DebateLabKIT/Phi-4-Argunaut-1-HIRPO with Docker Model Runner:
```
docker model run hf.co/DebateLabKIT/Phi-4-Argunaut-1-HIRPO
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for Phi-4-Argunaut-1-HIRPO

This model is a fine-tuned version of DebateLabKIT/Phi-4-Argunaut-1-SPIN-dev1. It has been trained using TRL.

📘 HF Blog Article

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="DebateLabKIT/Llama-3.1-Argunaut-1-8B-HIRPO", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with Hindsight Instruction Relabeling Preference Optimization (HIRPO), an Online DPO version derived from The Wisdom of Hindsight Makes Language Models Better Instruction Followers.

More details about the training procedure can be found in the blog post.

We have released the preference pairs generated online as a separate dataset: DebateLabKIT/argunauts-hirpo-preferences.

Framework versions

TRL: 0.19.1
Transformers: 4.53.3
Pytorch: 2.4.1
Datasets: 3.1.0
Tokenizers: 0.21.4

Evaluation

As described in this article, Phi-4-Argunaut-1-HIRPO technically masters formal argument analysis but has lost general conversational abilities during one-sided training.

Citations

Cite HIR as:

@misc{zhang2023wisdomhindsightmakeslanguage,
      title={The Wisdom of Hindsight Makes Language Models Better Instruction Followers}, 
      author={Tianjun Zhang and Fangchen Liu and Justin Wong and Pieter Abbeel and Joseph E. Gonzalez},
      year={2023},
      eprint={2302.05206},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2302.05206}, 
}

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}

Downloads last month: 6

Safetensors

Model size

415k params

Tensor type

BF16

Model tree for DebateLabKIT/Phi-4-Argunaut-1-HIRPO

Base model

DebateLabKIT/Phi-4-Argunaut-1-SPIN-dev1

Finetuned

(1)

this model

Quantizations

2 models

Dataset used to train DebateLabKIT/Phi-4-Argunaut-1-HIRPO

Paper for DebateLabKIT/Phi-4-Argunaut-1-HIRPO

The Wisdom of Hindsight Makes Language Models Better Instruction Followers

Paper • 2302.05206 • Published Feb 10, 2023