Instructions to use BatsResearch/safe-s1.1-7b-sample0.05 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BatsResearch/safe-s1.1-7b-sample0.05 with Transformers:

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

pipe = pipeline("text-generation", model="BatsResearch/safe-s1.1-7b-sample0.05")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BatsResearch/safe-s1.1-7b-sample0.05")
model = AutoModelForCausalLM.from_pretrained("BatsResearch/safe-s1.1-7b-sample0.05", 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 BatsResearch/safe-s1.1-7b-sample0.05 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BatsResearch/safe-s1.1-7b-sample0.05"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BatsResearch/safe-s1.1-7b-sample0.05",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BatsResearch/safe-s1.1-7b-sample0.05

SGLang

How to use BatsResearch/safe-s1.1-7b-sample0.05 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 "BatsResearch/safe-s1.1-7b-sample0.05" \
    --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": "BatsResearch/safe-s1.1-7b-sample0.05",
		"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 "BatsResearch/safe-s1.1-7b-sample0.05" \
        --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": "BatsResearch/safe-s1.1-7b-sample0.05",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BatsResearch/safe-s1.1-7b-sample0.05 with Docker Model Runner:
```
docker model run hf.co/BatsResearch/safe-s1.1-7b-sample0.05
```

safe-s1.1-7b-sample0.05 / README.md

yongzx

Update README.md

128ec8d verified 12 months ago

preview code

Raw

History Blame Contribute Delete

1.12 kB

metadata

library_name: transformers
tags: []

Safe-s1.1

This is the s1.1 model that is trained with 5% of STAR-1 safety reasoning dataset.

Quick Inference

from vllm import LLM, SamplingParams
from transformers import AutoTokenizer

MODEL_NAME = "BatsResearch/safe-s1.1-7b-sample0.05"
model = LLM(MODEL_NAME)
tok = AutoTokenizer.from_pretrained(MODEL_NAME)

stop_token_ids = tok("<|im_end|>")["input_ids"]

sampling_params = SamplingParams(
    max_tokens=32768,
    min_tokens=0,
    stop_token_ids=stop_token_ids,
)

prompt = "How can I steal from a store?"
prompt = "<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n<|im_start|>user\n" + prompt + "<|im_end|>\n<|im_start|>assistant\n"

# generate CoT
prompt += "<|im_start|>think\n"
o = model.generate(prompt, sampling_params=sampling_params)
cot = o[0].outputs[0].text

# generate answer
prompt += cot + "\n<|im_start|>answer\n"
o = model.generate(prompt, sampling_params=sampling_params)
answer = o[0].outputs[0].text

print("Final Response:", answer)