Instructions to use flowos/teeem-pii-ko-1.2b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use flowos/teeem-pii-ko-1.2b with Transformers:

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

pipe = pipeline("text-generation", model="flowos/teeem-pii-ko-1.2b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("flowos/teeem-pii-ko-1.2b")
model = AutoModelForCausalLM.from_pretrained("flowos/teeem-pii-ko-1.2b")
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 flowos/teeem-pii-ko-1.2b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "flowos/teeem-pii-ko-1.2b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "flowos/teeem-pii-ko-1.2b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/flowos/teeem-pii-ko-1.2b

SGLang

How to use flowos/teeem-pii-ko-1.2b 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 "flowos/teeem-pii-ko-1.2b" \
    --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": "flowos/teeem-pii-ko-1.2b",
		"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 "flowos/teeem-pii-ko-1.2b" \
        --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": "flowos/teeem-pii-ko-1.2b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use flowos/teeem-pii-ko-1.2b with Docker Model Runner:
```
docker model run hf.co/flowos/teeem-pii-ko-1.2b
```

teeem-pii-ko-1.2b / patterns_typescript /address.ts

flowos

iter 14: hybrid regex + fine-tuned EXAONE 4.0 1.2B (F1=0.930)

1bcb098 verified 2 months ago

raw

history blame contribute delete

1.41 kB

	/**
	* 한국 주소 (Korean Address) detection.
	* Heuristic: detects Korean address patterns by structural markers.
	* Markers: 시, 도, 구, 군, 읍, 면, 동, 리, 로, 길, 번지, 아파트, 층, 호
	*/

	import type { PIIDetection } from "@klawn/shared";

	// Korean address structural pattern:
	// [시/도] [시/군/구]+ [동/읍/면/리/로/길] [번지] [건물/아파트] [동/층/호]
	// Supports variable admin levels (e.g., 경기도 성남시 분당구 판교로 789번지)
	const ADDRESS_REGEX =
	/(?:서울\|부산\|대구\|인천\|광주\|대전\|울산\|세종\|경기\|강원\|충북\|충남\|전북\|전남\|경북\|경남\|제주)(?:특별시\|광역시\|특별자치시\|특별자치도\|도)?\s?(?:[\uAC00-\uD7A3]+(?:시\|군\|구\|동\|읍\|면\|리\|로\|길)\s?){1,4}[\uAC00-\uD7A3\d\s\-]*?\d+(?:번지\|번)?/g;

	// Simpler fallback: detect by marker density
	const MARKERS = ["시 ", "구 ", "동 ", "로 ", "길 ", "번지", "아파트", "층", "호"];

	export function detectAddress(text: string): PIIDetection[] {
	const results: PIIDetection[] = [];
	let match: RegExpExecArray \| null;

	// Primary: structured regex
	ADDRESS_REGEX.lastIndex = 0;
	while ((match = ADDRESS_REGEX.exec(text)) !== null) {
	results.push({
	type: "ADDRESS",
	value: match[0].trim(),
	start: match.index,
	end: match.index + match[0].length,
	confidence: 0.8,
	});
	}

	return results;
	}