Instructions to use Infinity08/KAWK-500M-Korean-Instruct-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Infinity08/KAWK-500M-Korean-Instruct-v1 with Transformers:

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

pipe = pipeline("text-generation", model="Infinity08/KAWK-500M-Korean-Instruct-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Infinity08/KAWK-500M-Korean-Instruct-v1")
model = AutoModelForCausalLM.from_pretrained("Infinity08/KAWK-500M-Korean-Instruct-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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Infinity08/KAWK-500M-Korean-Instruct-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Infinity08/KAWK-500M-Korean-Instruct-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": "Infinity08/KAWK-500M-Korean-Instruct-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Infinity08/KAWK-500M-Korean-Instruct-v1

SGLang

How to use Infinity08/KAWK-500M-Korean-Instruct-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 "Infinity08/KAWK-500M-Korean-Instruct-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": "Infinity08/KAWK-500M-Korean-Instruct-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 "Infinity08/KAWK-500M-Korean-Instruct-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": "Infinity08/KAWK-500M-Korean-Instruct-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Infinity08/KAWK-500M-Korean-Instruct-v1 with Docker Model Runner:
```
docker model run hf.co/Infinity08/KAWK-500M-Korean-Instruct-v1
```

KAWK-500M-Korean-Instruct-v1 / README.md

Infinity08

Update KAWK 500M Instruct model card

459b87b verified about 1 month ago

2.79 kB

language:
  - ko
library_name: transformers
pipeline_tag: text-generation
datasets:
  - Infinity08/KAWK500M-Korean-SFT-v1
tags:
  - llama
  - korean
  - causal-lm
  - instruction-tuned
  - kawk

KAWK 500M Korean Instruct v1

한국어 중심 말뭉치로 처음부터 사전학습한 505M 파라미터 KAWK 베이스 모델에 한국어 대화·지시 데이터로 supervised fine-tuning(SFT)을 적용한 모델입니다.

모델 구조

아키텍처: Llama 계열 decoder-only Transformer
파라미터: 505,350,400
어휘: 한국어 SentencePiece Unigram 32,000
레이어 / hidden / MLP: 26 / 1,280 / 3,584
Attention / KV heads: 20 / 5 (GQA)
최대 문맥: 2,048토큰
입력·출력 임베딩 공유

학습

베이스 사전학습량: 한국어 중심 약 100억 토큰
SFT 데이터: Infinity08/KAWK500M-Korean-SFT-v1
데이터 revision: 0a8e914359a063b916cfd9a3ee068a38ddcc1f79
SFT: 2 epochs, 739 optimizer steps, 약 9,080만 packed-token capacity
Assistant 응답 토큰에만 loss 적용
유효 배치: 60 sequences × 2,048 tokens
Precision / GPU: BF16 / NVIDIA A100 80GB
Learning rate: 2e-5 → 2e-6 cosine decay
step 700 validation loss / perplexity: 1.9786 / 7.2327
전체 설정과 로그: training/

사용 예시

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

repo_id = "Infinity08/KAWK-500M-Korean-Instruct-v1"
tokenizer = AutoTokenizer.from_pretrained(repo_id, use_fast=False)
model = AutoModelForCausalLM.from_pretrained(
    repo_id,
    dtype=torch.bfloat16,
    device_map="auto",
)

messages = [{"role": "user", "content": "대한민국의 수도를 간단히 설명해줘."}]
prompt = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(
    **inputs,
    max_new_tokens=128,
    do_sample=True,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.05,
)
new_tokens = outputs[0, inputs["input_ids"].shape[1]:]
print(tokenizer.decode(new_tokens, skip_special_tokens=True))

대화 템플릿은 ### 지침:, ### 사용자:, ### 도우미: 역할 헤더를 사용하며 토크나이저의 chat_template.jinja에 포함되어 있습니다.

평가

KoBEST·KMMLU 0-shot 평가는 진행 중이며 완료 후 원시 결과와 요약을 evaluation/에 추가합니다.

한계

500M급 모델이므로 복잡한 추론, 정확한 사실 회상, 긴 지시 수행을 보장하지 않습니다. SFT 데이터의 편향이나 잘못된 답을 재현할 수 있으며, 사실성·안전성·개인정보 재현에 대한 포괄적 평가가 완료되지 않았습니다. 고위험 의사결정에 사용하지 마십시오.