skt / A.X-3.1-Light

A.X 3.1 Light

🤗 Models | 🖥️ Github

A.X 3.1 Light Highlights

A.X 3.1 Light (pronounced "A dot X") is a light weight LLM optimized for Korean-language understanding and enterprise deployment. This sovereign AI model was developed entirely in-house by SKT, encompassing model architecture, data curation, and training, all carried out on SKT’s proprietary supercomputing infrastructure, TITAN. The model was trained from scratch on a high-quality multilingual corpus comprising 1.65 trillion tokens , with a primary focus on the Korean language. With a strong emphasis on data quality, A.X 3.1 Light achieves Pareto-optimal performance among Korean LLMs relative to its training corpus size , enabling highly efficient and cost-effective compute usage .

• Authentic Korean Sovereign AI : A.X 3.1 Light was trained on a high-quality multilingual dataset—fully curated in-house—using SKT’s proprietary GPU infrastructure.
• Highly Efficient Multilingual LLM : A.X 3.1 Light demonstrates superior performance among open-source Korean LLMs, despite its relatively compact training size of 1.65 trillion tokens.
• Superior Korean Proficiency : A.X 3.1 Light achieved a score of 61.7 on the KMMLU : the leading benchmark for Korean-language evaluation and a Korean-specific adaptation of MMLU, outperforming other Korean-specified models.
• Deep Korean Understanding : A.X 3.1 Light obtained 27.43 on the KoBALT-700 : a benchmark for Korean advanced linguistic tasks, outperforming other Korean-specialized models.
• Efficient Token Usage : A.X 3.1 Light requires approximately 33% fewer tokens than GPT-4o to process equivalent Korean inputs, facilitating more cost-effective and computationally efficient inference.
• Long-Context Handling : A.X 3.1 Light supports up to 32,768 tokens .

Core Technologies

A.X 3.1 Light represents an efficient sovereign AI model , developed end-to-end by SKT, encompassing model architecture, data curation, infrastructure deployment, and optimization.

Model Architecture Specs

Model	# Params	# Layers	# KV-Heads	Hidden Dim	FFN Dim
A.X 3.1 Light	7B	32	32	4096	10880

High-Quality Data Pipeline & Strategic Mixture

• We collected and curated a training dataset comprising 20 trillion tokens sourced from diverse domains.
• The entire dataset was processed through SKT’s proprietary data pipeline, incorporating synthetic data generation and comprehensive quality filtering.
• For training A.X 3.1 Light, a total of 1.65 trillion tokens were utilized, comprising a Korean-focused multilingual corpus.

Pareto-Optimal Compute Efficiency

A.X 3.1 Light achieves 5 to 6 times lower computational cost compared to models with similar performance levels. Rigorous data curation and two-stage training with STEM-focused data enabled competitive performance at reduced FLOPs.

Benchmark Results

Benchmarks	A.X 3.1 Light	Kanana-1.5-8B	EXAONE-3.5-7.8B	Qwen2.5-7B	Qwen3-8B (w/o reasoning)
Knowledge	KMMLU	61.70	48.28	53.76	49.56	63.53
CLIcK	71.22	61.30	64.11	58.30	63.31
KoBALT	27.43	23.14	21.71	21.57	26.57
MMLU	66.95	68.82	72.20	75.40	82.89
General	Ko-MT-Bench	78.56	76.30	81.06	61.31	64.06
MT-Bench	74.38	77.60	83.50	79.37	65.69
Instruction Following	Ko-IFEval	70.04	69.96	65.01	60.73	73.39
IFEval	79.86	80.11	82.61	76.73	85.38
Math	HRM8K	41.70	30.87	31.88	35.13	52.50
MATH	70.14	59.28	63.20	65.58	71.48
Code	HumanEval+	73.78	76.83	76.83	74.39	77.44
MBPP+	61.64	67.99	64.29	68.50	62.17

🚀 Quickstart

with HuggingFace Transformers

• transformers>=4.46.0 or the latest version is required to use skt/A.X-3.1-Light

pip install transformers>=4.46.0

Example Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "skt/A.X-3.1-Light"
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
model.eval()
tokenizer = AutoTokenizer.from_pretrained(model_name)

messages = [
    {"role": "system", "content": "당신은 사용자가 제공하는 영어 문장들을 한국어로 번역하는 AI 전문가입니다."},
    {"role": "user", "content": "The first human went into space and orbited the Earth on April 12, 1961."},
]
input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)

with torch.no_grad():
    output = model.generate(
        input_ids,
        max_new_tokens=128,
        do_sample=False,
    )

len_input_prompt = len(input_ids[0])
response = tokenizer.decode(output[0][len_input_prompt:], skip_special_tokens=True)
print(response)
# Output:
# 1961년 4월 12일, 최초의 인간이 우주에 나가 지구를 궤도를 돌았습니다.

with vLLM

• vllm>=v0.6.4.post1 or the latest version is required to use tool-use feature

pip install vllm>=v0.6.4.post1
# if you don't want to activate tool-use feature, just commenting out below vLLM option
VLLM_OPTION="--enable-auto-tool-choice --tool-call-parser hermes"
vllm serve skt/A.X-3.1-Light $VLLM_OPTION

Example Usage

from openai import OpenAI

def call(messages, model):
    completion = client.chat.completions.create(
        model=model,
        messages=messages,
    )
    print(completion.choices[0].message)

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="api_key"
)
model = "skt/A.X-3.1-Light"
messages = [{"role": "user", "content": "에어컨 여름철 적정 온도는? 한줄로 답변해줘"}]
call(messages, model)
# Output:
# 에어컨 여름철 적정 온도는 24~26도입니다.

messages = [{"role": "user", "content": "What is the appropriate temperature for air conditioning in summer? Respond in a single sentence."}]
call(messages, model)
# Output:
# The appropriate temperature for air conditioning in summer is generally set between 24 to 26°C for optimal comfort and energy efficiency.

Examples for tool-use

from openai import OpenAI


def call(messages, model):
    completion = client.chat.completions.create(
        model=model,
        messages=messages,
        tools=tools
    )
    print(completion.choices[0].message)


client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="api_key"
)
model = "skt/A.X-3.1-Light"

calculate_discount = {
    "type": "function",
    "function": {
        "name": "calculate_discount",
        "description": "원가격과 할인율(퍼센트 단위)을 입력받아 할인된 가격을계산한다.",
        "parameters": {
            "type": "object",
            "properties": {
                "original_price": {
                    "type": "number",
                    "description": "상품의 원래 가격"
                },
                "discount_percentage": {
                    "type": "number",
                    "description": "적용할 할인율"
                }
            },
            "required": ["original_price", "discount_percentage"]
        }
    }
}
get_exchange_rate = {
    "type": "function",
    "function": {
        "name": "get_exchange_rate",
        "description": "두 통화 간의 환율을 가져온다.",
        "parameters": {
            "type": "object",
            "properties": {
                "base_currency": {
                    "type": "string",
                    "description": "The currency to convert from."
                },
                "target_currency": {
                    "type": "string",
                    "description": "The currency to convert to."
                }
            },
            "required": ["base_currency", "target_currency"]
        }
    }
}
tools = [calculate_discount, get_exchange_rate]

### Slot filling ###
messages = [{"role": "user", "content": "우리가 뭘 사야되는데 원가가 57600원인데 직원할인 받으면 얼마야?"}]
call(messages, model)
# Output:
# ChatCompletionMessage(content='직원 할인을 적용하기 위해서는 할인율을 알 수 있어야 합니다. 할인율을 알려주실 수 있나요?', role='assistant', function_call=None, tool_calls=[], reasoning_content=None)


### Function calling ###
messages = [
    {"role": "user", "content": "우리가 뭘 사야되는데 원가가 57600원인데 직원할인 받으면 얼마야?"},
    {"role": "assistant", "content": "직원 할인을 적용하기 위해서는 할인율을 알 수 있어야 합니다. 할인율을 알려주실 수 있나요?"},
    {"role": "user", "content": "15% 할인 받을 수 있어."},
]
call(messages, model)
# Output: 
# ChatCompletionMessage(content=None, role='assistant', function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='chatcmpl-tool-3ebf11847364450daf363039db80cc50', function=Function(arguments='{"original_price": 57600, "discount_percentage": 15}', name='calculate_discount'), type='function')], reasoning_content=None)


### Completion ###
messages = [
    {"role": "user", "content": "우리가 뭘 사야되는데 원가가 57600원인데 직원할인 받으면 얼마야?"},
    {"role": "assistant", "content": ""},
    {"role": "user", "content": "15% 할인 받을 수 있어."},
    {"role": "tool", "tool_call_id": "random_id", "name": "calculate_discount", "content": "{\"original_price\": 57600, \"discount_percentage\": 15, \"discounted_price\": 48960.0}"}
]
call(messages, model)
# Output: 
# ChatCompletionMessage(content='57,600원의 상품에 15% 할인을 적용하면, 할인된 가격은 48,960원입니다.', role='assistant', function_call=None, tool_calls=[], reasoning_content=None)

License

The A.X 3.1 Light model is licensed under Apache License 2.0 .

Citation

@article{SKTAdotX3.1Light,
  title={A.X 3.1 Light},
  author={SKT AI Model Lab},
  year={2025},
  url={https://huggingface.co/skt/A.X-3.1-Light}
}

Contact

• Business & Partnership Contact: [email protected]