Instructions to use Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 with Transformers:

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

pipe = pipeline("text-generation", model="Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1")
model = AutoModelForCausalLM.from_pretrained("Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1

SGLang

How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 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 "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1" \
    --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": "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1",
		"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 "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1" \
        --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": "Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 with Docker Model Runner:
```
docker model run hf.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1
```

Qwen3-8B-Base-RAQUEL-MUSE-Unlearn-NPO-GD-LoRA-v1 / evaluation.json

Hyukkyu

RAQUEL2 LoRA release (muse_qwen3_npo, data ac827565)

164576a verified 15 days ago

Raw History Blame Contribute Delete

3.12 kB

	{
	"data_repository": "Hyukkyu/RAQUEL2-ICLR",
	"data_revision": "ac82756570fcce84441fb413ab523de8da679efd",
	"evaluated_adapter": "early_stop",
	"evidence": {
	"judge_inputs_sha256": "a15e558b21eaf717af7fec76d2662695d19a280abed58b4e8e8a324ff9e3ea59",
	"judgments_sha256": "beac9a8026b7f85cbbf8236f650b7dbee5fb7f514ca27f23c00def1164a890f6"
	},
	"generation": {
	"do_sample": false,
	"max_new_tokens": 96,
	"max_prompt_length": 512,
	"prompt": "Question: {question}\nAnswer:"
	},
	"judge": {
	"model": "Qwen/Qwen3.8-27B",
	"served_name": "qwen3.8-27b",
	"server": "vLLM 0.28.0",
	"temperature": 0,
	"thinking": false
	},
	"merge_audit": {
	"max_kl": 0.030766041949391365,
	"maximum_allowed_kl": 0.1,
	"mean_absolute_logit_difference": 0.15677118301391602,
	"merge_algorithm": "fp32_sum_then_cast_once",
	"prompts": 16,
	"top1_agreement": 1.0
	},
	"metrics": {
	"muse_forget": {
	"accuracy": 0.096,
	"correct": 72,
	"total": 750
	},
	"muse_forget_rephrased": {
	"accuracy": 0.11605937921727395,
	"correct": 86,
	"total": 741
	},
	"muse_retain": {
	"accuracy": 0.9676923076923077,
	"correct": 1258,
	"total": 1300
	},
	"muse_retain_rephrased": {
	"accuracy": 0.4770070148090413,
	"correct": 612,
	"total": 1283
	},
	"raquel_affected": {
	"accuracy": 0.1304856641310708,
	"correct": 223,
	"total": 1709
	},
	"raquel_unaffected": {
	"accuracy": 0.14029850746268657,
	"correct": 282,
	"total": 2010
	}
	},
	"reference": {
	"M_orig": {
	"muse_forget": {
	"accuracy": 0.9986666666666667,
	"correct": 749,
	"total": 750
	},
	"muse_forget_rephrased": {
	"accuracy": 0.8097165991902834,
	"correct": 600,
	"total": 741
	},
	"muse_retain": {
	"accuracy": 0.9992307692307693,
	"correct": 1299,
	"total": 1300
	},
	"muse_retain_rephrased": {
	"accuracy": 0.7583787996882307,
	"correct": 973,
	"total": 1283
	},
	"raquel_affected": {
	"accuracy": 0.19777647747220597,
	"correct": 338,
	"total": 1709
	},
	"raquel_unaffected": {
	"accuracy": 0.20696517412935322,
	"correct": 416,
	"total": 2010
	}
	},
	"M_ret": {
	"muse_forget": {
	"accuracy": 0.17333333333333334,
	"correct": 130,
	"total": 750
	},
	"muse_forget_rephrased": {
	"accuracy": 0.1659919028340081,
	"correct": 123,
	"total": 741
	},
	"muse_retain": {
	"accuracy": 1.0,
	"correct": 1300,
	"total": 1300
	},
	"muse_retain_rephrased": {
	"accuracy": 0.7622759158222915,
	"correct": 978,
	"total": 1283
	},
	"raquel_affected": {
	"accuracy": 0.12521942656524282,
	"correct": 214,
	"total": 1709
	},
	"raquel_unaffected": {
	"accuracy": 0.12985074626865672,
	"correct": 261,
	"total": 2010
	}
	}
	}
	}