Instructions to use Jiajunruan/NPO-Fix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jiajunruan/NPO-Fix with Transformers:

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

pipe = pipeline("text-generation", model="Jiajunruan/NPO-Fix")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jiajunruan/NPO-Fix")
model = AutoModelForCausalLM.from_pretrained("Jiajunruan/NPO-Fix")
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 Jiajunruan/NPO-Fix with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Jiajunruan/NPO-Fix

SGLang

How to use Jiajunruan/NPO-Fix 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 "Jiajunruan/NPO-Fix" \
    --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": "Jiajunruan/NPO-Fix",
		"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 "Jiajunruan/NPO-Fix" \
        --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": "Jiajunruan/NPO-Fix",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Jiajunruan/NPO-Fix with Docker Model Runner:
```
docker model run hf.co/Jiajunruan/NPO-Fix
```

Add library_name, pipeline_tag and GitHub link

by nielsr HF Staff - opened May 29

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-13

Files changed (1) hide show

README.md +8 -13

README.md CHANGED Viewed

@@ -1,10 +1,13 @@
 ---
-datasets:
-- locuslab/TOFU
 base_model:
 - meta-llama/Llama-3.2-1B-Instruct
 tags:
-- Unlearning, Forget10
 ---
 # **NPO-Fix:** An enhancement of NPO method with self-generated dataset for robust unlearning under probabilistic decoding.
@@ -15,20 +18,13 @@ tags:
 - **Base Method:** NPO.
 - **Original Model:** [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct).
 ### Model Sources
-<!-- Provide the basic links for the model. -->
 - **Paper:** [Leak@k: Unlearning Does Not Make LLMs Forget Under Probabilistic Decoding](https://arxiv.org/abs/2511.04934)
 ## Citation
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
 **BibTeX:**
 ```bibtex
@@ -42,5 +38,4 @@ tags:
 ## Model Card Authors
-[Jiajun Ruan: jruan@umn.edu]

 ---
 base_model:
 - meta-llama/Llama-3.2-1B-Instruct
+datasets:
+- locuslab/TOFU
 tags:
+- Unlearning
+- Forget10
+library_name: transformers
+pipeline_tag: text-generation
 ---
 # **NPO-Fix:** An enhancement of NPO method with self-generated dataset for robust unlearning under probabilistic decoding.
 - **Base Method:** NPO.
 - **Original Model:** [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct).
 ### Model Sources
 - **Paper:** [Leak@k: Unlearning Does Not Make LLMs Forget Under Probabilistic Decoding](https://arxiv.org/abs/2511.04934)
+- **Repository:** https://github.com/OptimAI-Lab/Leak-k
 ## Citation
 **BibTeX:**
 ```bibtex
 ## Model Card Authors
+[Jiajun Ruan: jruan@umn.edu]