---
license: apache-2.0
language:
- ko
base_model:
- Qwen/Qwen2.5-14B-Instruct
---
# Announcing OLAFv2: The Next Step in Korean Language Understanding ๐
We are thrilled to announce the release of **OLAFv2**, our state-of-the-art Korean language model, now available on Hugging Face! ๐ Designed to excel in complex reasoning, mathematical problem-solving, and general language understanding, OLAFv2 represents a significant leap forward in NLP capabilities for the Korean language.
## Key Features of OLAFv2 ๐
### **Two Model Sizes for Flexibility**
OLAFv2 is available in two parameter sizes:
- **14B (Billion) Parameters**: For maximum performance. ๐๏ธโโ๏ธ
- **1.5B (Billion) Parameters**: For lightweight applications and hardware-constrained environments. ๐ชถ
### **Reasoning Mode for Complex Tasks** ๐ค
One of OLAFv2's standout features is its **Reasoning Mode**, specifically designed for:
- Complex mathematical problem-solving. โ๏ธโ
- STEM (Science, Technology, Engineering, Mathematics) applications. ๐ฌ๐
- Tasks requiring detailed step-by-step reasoning. ๐ง
This mode can be effectively utilized for **Test-Time Scaling**, enabling the model to harness additional computational resources during inference. This approach enhances output detail and accuracy, achieving performance levels that surpass GPT-4o. ๐
### **Long Context Support** ๐
With support for up to **32K tokens**, OLAFv2 is perfect for:
- Retrieval-Augmented Generation (RAG). ๐ ๏ธ
- Tasks requiring long-context understanding and reasoning. ๐งต
## Benchmarks and Performance ๐
We share evaluation results across three benchmarks, KMMLU, HRM8K and LogicKor.
We also share results with inference-time scaling. For more details have a look into our [blog](https://www.onelineai.com/blog/test-time-scaling).
## Getting Started ๐
OLAFv2 is now available on Hugging Face! You can start using it by accessing our repository:
```python
# pip install transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "OLAResearch/OLAF2-14B"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "introduce yourself!"
messages = [
{"role": "system", "content": "You're name is OLAF. A large language model made by OneLineAI, specializing in Korean culture and finance."},
# for reasoning mode
#{"role": "system", "content": "You're name is OLAF. A large language model made by OneLineAI, specializing in Korean culture and finance.Perform two-step reasoning. Return your answers in \\boxed{N} format."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
