Instructions to use Yunhao-Feng/AdaGuard-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Yunhao-Feng/AdaGuard-8B with Transformers:

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

pipe = pipeline("text-generation", model="Yunhao-Feng/AdaGuard-8B")
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("Yunhao-Feng/AdaGuard-8B")
model = AutoModelForCausalLM.from_pretrained("Yunhao-Feng/AdaGuard-8B", 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 Yunhao-Feng/AdaGuard-8B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Yunhao-Feng/AdaGuard-8B

SGLang

How to use Yunhao-Feng/AdaGuard-8B 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 "Yunhao-Feng/AdaGuard-8B" \
    --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": "Yunhao-Feng/AdaGuard-8B",
		"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 "Yunhao-Feng/AdaGuard-8B" \
        --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": "Yunhao-Feng/AdaGuard-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Yunhao-Feng/AdaGuard-8B with Docker Model Runner:
```
docker model run hf.co/Yunhao-Feng/AdaGuard-8B
```

AdaGuard-8B / evaluation_results.json

Yunhao-Feng

Release AdaGuard weights, model card, and usage examples

41a51b7 verified 9 days ago

Raw History Blame Contribute Delete

1.54 kB

	{
	"schema_version": 1,
	"model_id": "Yunhao-Feng/AdaGuard-8B",
	"units": "percent",
	"result_type": "previously_reported_single_run",
	"release_validation_is_benchmark_rerun": false,
	"datasets": {
	"AdaptiveSafety": {
	"samples": 1000,
	"compliant": 500,
	"violating": 500,
	"binary": {
	"accuracy": 89.5,
	"precision": 96.25,
	"recall": 82.2,
	"f1": 88.67
	},
	"rule_identification": {
	"exact_match": 77.1,
	"micro_precision": 85.01,
	"micro_recall": 65.59,
	"micro_f1": 74.05
	}
	},
	"DynaBench": {
	"samples": 543,
	"compliant": 276,
	"violating": 267,
	"binary": {
	"accuracy": 76.8,
	"precision": 77.22,
	"recall": 74.91,
	"f1": 76.05
	},
	"rule_identification": {
	"exact_match": 70.72,
	"micro_precision": 61.11,
	"micro_recall": 65.92,
	"micro_f1": 63.42
	}
	}
	},
	"protocol": {
	"inference_dtype": "bfloat16",
	"decoding": "greedy",
	"prompt_budget_tokens": 16000,
	"max_new_tokens": 512,
	"hardware": "8 x NVIDIA A100-SXM4-80GB; one model replica per GPU",
	"input_truncation_recorded": false,
	"binary_positive_class": "violation",
	"invalid_binary": "counted as an error",
	"invalid_exact_match": "never credited",
	"invalid_rule_micro": "empty predicted rule set; gold violations remain false negatives",
	"rule_order": "same order as supplied policy; incorrect ordering is invalid"
	}
	}