Instructions to use mzhaoshuai/zephyr-7b-alpha-conf-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mzhaoshuai/zephyr-7b-alpha-conf-sft with Transformers:

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

pipe = pipeline("text-generation", model="mzhaoshuai/zephyr-7b-alpha-conf-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mzhaoshuai/zephyr-7b-alpha-conf-sft")
model = AutoModelForCausalLM.from_pretrained("mzhaoshuai/zephyr-7b-alpha-conf-sft", device_map="auto")
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 mzhaoshuai/zephyr-7b-alpha-conf-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mzhaoshuai/zephyr-7b-alpha-conf-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mzhaoshuai/zephyr-7b-alpha-conf-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mzhaoshuai/zephyr-7b-alpha-conf-sft

SGLang

How to use mzhaoshuai/zephyr-7b-alpha-conf-sft 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 "mzhaoshuai/zephyr-7b-alpha-conf-sft" \
    --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": "mzhaoshuai/zephyr-7b-alpha-conf-sft",
		"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 "mzhaoshuai/zephyr-7b-alpha-conf-sft" \
        --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": "mzhaoshuai/zephyr-7b-alpha-conf-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mzhaoshuai/zephyr-7b-alpha-conf-sft with Docker Model Runner:
```
docker model run hf.co/mzhaoshuai/zephyr-7b-alpha-conf-sft
```

RefAlign: RL with Similarity-based Rewards

GitHub repository: https://github.com/mzhaoshuai/RefAlign

Paper: Learning from Reference Answers: Versatile Language Model Alignment without Binary Human Preference Data.

This repository contains the SFT (Supervised Fine-Tuning) model mzhaoshuai/zephyr-7b-alpha-conf-sft, which is an integral part of the RefAlign framework. This model serves as an initial SFT step for Confidence Alignment experiments, trained with shuchangtao/CONQORD_dataset (specifically, conqord_step1_data), as described in the accompanying research.

Abstract

Large language models~(LLMs) are expected to be helpful, harmless, and honest. In different alignment scenarios, such as safety, confidence, and general preference alignment, binary preference data collection and reward modeling are resource-intensive but play a central role in transferring human preferences. In this work, we explore using the similarity between sampled generations and reference answers as a supplementary reward function for alignment. When unary reference answers are available, such similarity-based rewards can circumvent the need for binary preference data and explicit reward modeling. We introduce \textit{RefAlign}, a versatile REINFORCE-style alignment algorithm that does not rely on reward or reference models. RefAlign utilizes language generation evaluation metrics, such as BERTScore, between sampled generations and reference answers as surrogate rewards. Beyond general preference optimization, RefAlign can be naturally extended to diverse scenarios, including safety and confidence alignment, by combining similarity-based rewards with task-specific objectives. Across multiple scenarios, RefAlign achieves performance comparable to prior alignment methods while operating without binary preference data or reward models.

Framework versions

PEFT 0.11.1
Transformers 4.40.0

Bibtex

@article{zhao2025learning,
  title={Learning from reference answers: Versatile language model alignment without binary human preference data},
  author={Zhao, Shuai and Xu, Yunqiu and Zhu, Linchao and Yang, Yi},
  journal={arXiv preprint arXiv:2504.09895},
  year={2025}
}