KISTI-KONI
/

KONI-7B-R-20250831

@@ -1,199 +1,331 @@
 ---
 library_name: transformers
-tags: []
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
+license: apache-2.0
+language:
+- ko
+- en
+tags:
+- text-generation
+- pytorch
+- A.X 3
+- KISTI
+- KONI
+- 7b
 library_name: transformers
+base_model:
+- skt/A.X-3.1-Light
 ---
+# KONI-7B-R-20250831
+[**KONI (KISTI Open Neural Intelligence)**](https://huggingface.co/KISTI-KONI) is a large language model developed by the Korea Institute of Science and Technology Information (KISTI). Designed specifically for the scientific and technological domains, KONI excels in both Korean and English, making it an ideal tool for tasks requiring specialized knowledge in these areas.
+<div class="logo-row">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/RJUQtsIG1xwSHk2KuuDl_.png" alt="Koni logo" class="koni-logo">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/xYJqy-qqWzN2FjQwcV80Z.png" alt="OneLineAI logo" class="OneLineAI-logo">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/mJjX11ZN6dYMihNAk_IFn.png" alt="HAE-RAE logo" class="HAE-RAE-logo">
+</div>
+<style>
+  .logo-row {
+    display: flex;
+    align-items: center;      /* vertical align */
+    justify-content: center;  /* center the pair (optional) */
+    gap: 3px;                /* space between logos */
+    flex-wrap: wrap;          /* stack on very small screens */
+  }
+  .logo-row img {
+    width: auto;    /* keep aspect ratio */
+    object-fit: contain;
+  }
+  .koni-logo {
+    height: 200px;   /* KONI 크게 */
+  }
+  .OneLineAI-logo {
+    height: 110px;   /* KONI 크게 */
+  }
+  .HAE-RAE-logo {
+    height: 160px;   /* KONI 크게 */
+  }
+</style>
+**KONI-7B-R-20250831** is an 7B Korean reasoning–specialized model developed collaboratively by KISTI, OneLineAI, HAE-RAE, and ORACLE as part of the [**KO-REAson series**](https://huggingface.co/KOREAson). Built upon [A.X-3.1-Light](https://huggingface.co/skt/A.X-3.1-Light), it is an instruction-tuned variant optimized for Korean-centric reasoning with full English support. By leveraging the **Language-Mixed Chain-of-Thought strategy**—interleaving Korean and English during the reasoning stage—the model improves both consistency and accuracy in complex reasoning. It is designed to address a wide range of tasks, from science and technology queries to general knowledge, mathematics, and logical problem-solving. While fully maintaining the tokenizer, context length, and API compatibility of the base model, it further enhances performance through supervised fine-tuning (SFT) tailored for Korean reasoning and terminology preservation.
+## Key Features
+- **Korean-Centric Reasoning with English Support**: Optimized primarily for Korean reasoning tasks while providing full support for English, enabling robust bilingual usage.
+- **Language-Mixed Chain-of-Thought**: Employs Language-Mixed Chain-of-Thought strategy that interleaves Korean and English during the thought process, improving both consistency and accuracy in complex reasoning.
+- **Specialized in Science & Technology**: Trained with strong emphasis on scientific and technological domains, making it well-suited for expert-level queries in these areas.
+- **Base Model**: Built upon [A.X-3.1-Light](https://huggingface.co/skt/A.X-3.1-Light).
+- **Alignment**: Enhanced through Supervised Fine-Tuning (SFT) on 260k Language-Mixed Chain-of-Thought (CoT) examples, tailored for bilingual(Korean/English) reasoning and terminology preservation.
+- **Strengths**: Demonstrating substantial performance gains across diverse reasoning benchmarks, this model provides coherent and complex reasoning in both Korean and English, capable of addressing a broad spectrum of tasks such as science and technology queries, general knowledge, mathematics, and logical problem-solving.
+- **Intended Use**: Designed for science and technology Q&A, mathematical and logical problem-solving, Korean document understanding, and as a reasoning backbone for agent systems.
+---
+# KO-REAson
+[**KO-REAson**](https://huggingface.co/KOREAson) is a series of Korean-centric reasoning language models developed in collaboration with [OneLineAI](https://onelineai.com/), [KISTI-KONI](https://huggingface.co/KISTI-KONI), [HAE-RAE](https://huggingface.co/HAERAE-HUB) and ORACLE.
+We use the **Language-Mixed Chain-of-Thought (CoT)** approach, which allows the model to alternate between English and Korean during the “Think” stage of reasoning, preserving key Korean terms while leveraging English for logical scaffolding.
+Top-performing models of our series [KONI-7B-R-20250831 (KO-REAson-AX3_1-7B-0831)](https://huggingface.co/KISTI-KONI/KONI-7B-R-20250831) and [KO-REAson-7B-Q2_5-0831](https://huggingface.co/KoReason/KO-REASon-7B-Q2_5-0831) show performance comparable to models trained on closed-source datasets such as Exaone-Deep-7.8B.
+<p align="center">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/uqrKdxbQEqAFknYBmuH7Y.png"
+       alt="Model Comparison" width="500"/>
+  <em style="display:inline-block; text-align:center; white-space:normal; word-wrap:break-word; line-height:1.5; margin-top:0px;">
+    Average performance (Held-out-Ko) of open models trained on closed or open data. <br>
+    (Our models are highlighted in green.)
+  </em>
+</p>
+---
+## Model Details
+The **KO-REAson-0831** family comes in six variants based on the base model used.
+<div align="center">
+| Model                                                                                  | Base                 |
+| :--------------------------------------------------------------------------------------------: | :--------------------: |
+| [KO-REAson-L3_1-8B-0831](https://huggingface.co/KoReason/KO-REASon-L3_1-8B-0831)   | [Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct)      |
+| [KONI-Llama3.1-8B-R-20250831(KO-REAson-KL3_1-8B-0831)](https://huggingface.co/KISTI-KONI/KONI-Llama3.1-8B-R-20250831) |  [KONI-Llama3.1-8B-Instruct-20241024](https://huggingface.co/KISTI-KONI/KONI-Llama3.1-8B-Instruct-20241024)   |
+| [KO-REAson-G3-4B-0831](https://huggingface.co/KoReason/KO-REASon-G3-4B-0831)        | [Gemma-3-4b-it](https://huggingface.co/google/gemma-3-4b-it)           |
+| [KONI-7B-R-20250831(KO-REAson-AX3_1-7B-0831)](https://huggingface.co/KISTI-KONI/KONI-7B-R-20250831) |  [A.X-3.1-Light (≈7B)](https://huggingface.co/skt/A.X-3.1-Light)  |
+| [KO-REAson-K2505_8B-0831](https://huggingface.co/KoReason/KO-REASon-K2505_8B-0831) | [kanana-1.5-8b-instruct-2505](https://huggingface.co/kakaocorp/kanana-1.5-8b-instruct-2505)     |
+| [KO-REAson-7B-Q2_5-0831](https://huggingface.co/KoReason/KO-REASon-7B-Q2_5-0831)   | [Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct)        |
+</div>
+---
+# Performance
+**Evaluation Datasets**
+The model's performance was evaluated across a total of 11 benchmarks, and the evaluation suite is divided into two parts: (You can check these benchmarks in [HAERAE-HUB/KoSimpleEval](https://huggingface.co/datasets/HAERAE-HUB/KoSimpleEval))
+- **Held-in**: This set of benchmarks is used for routine monitoring of the model's performance during the training and ablation study phases.
+- **Held-out**: This set is used only once to evaluate the final model after all training and ablations are complete.
+This separation is designed to prevent inadvertent overfitting to the benchmarks during the iterative training process and to provide a more accurate measure of the model's generalization capabilities.
+<div align="center">
+|**Category**|**Held-in**|**Held-out**|
+|:---:|:---:|:---:|
+|**General Knowledge**|KMMLU-Redux|KMMLU-HARD, KMMLU-Pro|
+|**Reasoning**|MCLM|KSM, GPQA, AIME2024, AIME2025|
+|**Korean-specific**|HAE-RAE Bench|CLIcK, KoBALT-700|
+</div>
+**Comparison with models trained on public datasets**
+<table align="center" style="margin:auto; border-collapse:collapse; text-align:center; vertical-align:middle;">
+  <thead>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">Models</th>
+      <th style="text-align:center; vertical-align:middle;">#Instances</th>
+      <th style="text-align:center; vertical-align:middle;">Methodology</th>
+      <th style="text-align:center; vertical-align:middle;">Held-Out(Ko)</th>
+      <th style="text-align:center; vertical-align:middle;">Held-Out(En)</th>
+      <th style="text-align:center; vertical-align:middle;">Total</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">KONI-7B-R-20250831<br>(KO-REASon-AX3_1-7B-0831; Ours)</th>
+      <td style="text-align:center; vertical-align:middle;">260k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;"><b>44.60</b></td>
+      <td style="text-align:center; vertical-align:middle;">41.20</td>
+      <td style="text-align:center; vertical-align:middle;"><u>43.30</u></td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">KONI-Llama3.1-8B-R-20250831<br>(KO-REAson-KL3_1-8B-0831; Ours)</th>
+      <td style="text-align:center; vertical-align:middle;">260k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;">40.13</td>
+      <td style="text-align:center; vertical-align:middle;">30.57</td>
+      <td style="text-align:center; vertical-align:middle;">43.66</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">KO-REASon-7B-Q2_5-0831<br>(Ours)</th>
+      <td style="text-align:center; vertical-align:middle;">260k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;"><b>45.10</b></td>
+      <td style="text-align:center; vertical-align:middle;">38.75</td>
+      <td style="text-align:center; vertical-align:middle;"><u>49.95</u></td>
+    </tr>
+    <tr>
+      <td colspan="6" style="text-align:center; font-weight:bold;">Open Recipe (En)</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">OpenThinker3-7B</th>
+      <td style="text-align:center; vertical-align:middle;">1.2M</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;">33.60</td>
+      <td style="text-align:center; vertical-align:middle;"><b>55.50</b></td>
+      <td style="text-align:center; vertical-align:middle;">41.80</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">s1.1-7B</th>
+      <td style="text-align:center; vertical-align:middle;">1k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;">35.60</td>
+      <td style="text-align:center; vertical-align:middle;">23.40</td>
+      <td style="text-align:center; vertical-align:middle;">31.10</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">Llama-3.1-Nemotron-Nano-8B-v1</th>
+      <td style="text-align:center; vertical-align:middle;">&gt;3M</td>
+      <td style="text-align:center; vertical-align:middle;">SFT &amp; RL</td>
+      <td style="text-align:center; vertical-align:middle;">27.00</td>
+      <td style="text-align:center; vertical-align:middle;">44.10</td>
+      <td style="text-align:center; vertical-align:middle;">33.40</td>
+    </tr>
+    <tr>
+      <td colspan="6" style="text-align:center; font-weight:bold;">Open Recipe (Ko)</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">Ko-R1-14B</th>
+      <td style="text-align:center; vertical-align:middle;">45k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;"><u>43.70</u></td>
+      <td style="text-align:center; vertical-align:middle;"><u>46.30</u></td>
+      <td style="text-align:center; vertical-align:middle;"><b>44.70</b></td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">Ko-R1-7B</th>
+      <td style="text-align:center; vertical-align:middle;">45k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;">27.30</td>
+      <td style="text-align:center; vertical-align:middle;">36.10</td>
+      <td style="text-align:center; vertical-align:middle;">30.60</td>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">LLaMa-3.1-Ko-Reasoning-8B</th>
+      <td style="text-align:center; vertical-align:middle;">63k</td>
+      <td style="text-align:center; vertical-align:middle;">SFT</td>
+      <td style="text-align:center; vertical-align:middle;">17.70</td>
+      <td style="text-align:center; vertical-align:middle;">7.70</td>
+      <td style="text-align:center; vertical-align:middle;">14.00</td>
+    </tr>
+  </tbody>
+</table>
+<br>
+**Held-out benchmark performance**
+<table border="1" cellspacing="0" cellpadding="6" style="margin:auto; border-collapse:collapse; text-align:center; vertical-align:middle; white-space:nowrap;">
+  <thead>
+    <tr>
+      <th rowspan="2">Model</th>
+      <th rowspan="2">Model Size</th>
+      <th colspan="2">General</th>
+      <th colspan="4">Reasoning</th>
+      <th colspan="2">Korean-Specific</th>
+      <th rowspan="2">Average<br>(Held-out)</th>
+      <th rowspan="2">Average<br>(Held-out-Ko)</th>
+    </tr>
+    <tr>
+      <th style="text-align:center; vertical-align:middle;">KMMLU-HARD</th>
+      <th style="text-align:center; vertical-align:middle;">KMMLU-Pro</th>
+      <th style="text-align:center; vertical-align:middle;">KSM</th>
+      <th style="text-align:center; vertical-align:middle;">AIME 2024</th>
+      <th style="text-align:center; vertical-align:middle;">AIME 2025</th>
+      <th style="text-align:center; vertical-align:middle;">GPQA</th>
+      <th style="text-align:center; vertical-align:middle;">CLIcK</th>
+      <th style="text-align:center; vertical-align:middle;">KoBALT-700</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>Llama-3.1-Nemotron-Nano-8B</b></td>
+      <td style="text-align:center; vertical-align:middle;">8.03</td><td style="text-align:center; vertical-align:middle;">21.47</td><td style="text-align:center; vertical-align:middle;">22.89</td><td style="text-align:center; vertical-align:middle;">47.06</td><td style="text-align:center; vertical-align:middle;">56.67</td><td style="text-align:center; vertical-align:middle;">43.33</td><td style="text-align:center; vertical-align:middle;">32.32</td><td style="text-align:center; vertical-align:middle;">34.54</td><td style="text-align:center; vertical-align:middle;">9.29</td><td style="text-align:center; vertical-align:middle;">33.45</td><td style="text-align:center; vertical-align:middle;">27.05</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>Llama-3.1-Korean-Reasoning-8B-Instruct</b></td>
+      <td style="text-align:center; vertical-align:middle;">8.03</td><td style="text-align:center; vertical-align:middle;">14.91</td><td style="text-align:center; vertical-align:middle;">21.72</td><td style="text-align:center; vertical-align:middle;">6.09</td><td style="text-align:center; vertical-align:middle;">0.00</td><td style="text-align:center; vertical-align:middle;">0.00</td><td style="text-align:center; vertical-align:middle;">23.23</td><td style="text-align:center; vertical-align:middle;">39.65</td><td style="text-align:center; vertical-align:middle;">6.14</td><td style="text-align:center; vertical-align:middle;">13.97</td><td style="text-align:center; vertical-align:middle;">17.70</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>EXAONE-Deep-7.8B</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.82</td><td style="text-align:center; vertical-align:middle;"><u>40.96</u></td><td style="text-align:center; vertical-align:middle;">37.35</td><td style="text-align:center; vertical-align:middle;"><b>70.80</b></td><td style="text-align:center; vertical-align:middle;"><b>70.00</b></td><td style="text-align:center; vertical-align:middle;"><b>63.33</b></td><td style="text-align:center; vertical-align:middle;"><b>64.65</b></td><td style="text-align:center; vertical-align:middle;">54.24</td><td style="text-align:center; vertical-align:middle;">18.86</td><td style="text-align:center; vertical-align:middle;"><b>52.52</b></td><td style="text-align:center; vertical-align:middle;">44.44</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>DeepSeek-R1-Distill-Qwen-7B</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.62</td><td style="text-align:center; vertical-align:middle;">0.00</td><td style="text-align:center; vertical-align:middle;">23.00</td><td style="text-align:center; vertical-align:middle;">56.09</td><td style="text-align:center; vertical-align:middle;">60.00</td><td style="text-align:center; vertical-align:middle;">40.00</td><td style="text-align:center; vertical-align:middle;">43.43</td><td style="text-align:center; vertical-align:middle;">0.00</td><td style="text-align:center; vertical-align:middle;">8.29</td><td style="text-align:center; vertical-align:middle;">28.85</td><td style="text-align:center; vertical-align:middle;">17.48</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>DeepSeek-R1-Distill-Llama-8B</b></td>
+      <td style="text-align:center; vertical-align:middle;">8.03</td><td style="text-align:center; vertical-align:middle;">23.22</td><td style="text-align:center; vertical-align:middle;">26.26</td><td style="text-align:center; vertical-align:middle;">29.97</td><td style="text-align:center; vertical-align:middle;">33.33</td><td style="text-align:center; vertical-align:middle;">20.00</td><td style="text-align:center; vertical-align:middle;"><U>46.46</u></td><td style="text-align:center; vertical-align:middle;">39.05</td><td style="text-align:center; vertical-align:middle;">13.29</td><td style="text-align:center; vertical-align:middle;">28.95</td><td style="text-align:center; vertical-align:middle;">26.36</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>s1.1-7B</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.62</td><td style="text-align:center; vertical-align:middle;">31.16</td><td style="text-align:center; vertical-align:middle;"><u>37.70</u></td><td style="text-align:center; vertical-align:middle;">30.60</td><td style="text-align:center; vertical-align:middle;">16.67</td><td style="text-align:center; vertical-align:middle;">23.33</td><td style="text-align:center; vertical-align:middle;">30.30</td><td style="text-align:center; vertical-align:middle;"><u>56.84</u></td><td style="text-align:center; vertical-align:middle;"><u>21.86</u></td><td style="text-align:center; vertical-align:middle;">31.06</td><td style="text-align:center; vertical-align:middle;">35.63</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>OpenThinker3-7B</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.62</td><td style="text-align:center; vertical-align:middle;">30.31</td><td style="text-align:center; vertical-align:middle;">26.26</td><td style="text-align:center; vertical-align:middle;"><u>63.59</u></td><td style="text-align:center; vertical-align:middle;"><u>66.67</u></td><td style="text-align:center; vertical-align:middle;"><u>53.33</u></td><td style="text-align:center; vertical-align:middle;"><u>46.46</u></td><td style="text-align:center; vertical-align:middle;">47.69</td><td style="text-align:center; vertical-align:middle;">10.14</td><td style="text-align:center; vertical-align:middle;">35.63</td><td style="text-align:center; vertical-align:middle;">30.60</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>Ko-R1-7B</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.61</td><td style="text-align:center; vertical-align:middle;">28.46</td><td style="text-align:center; vertical-align:middle;">19.31</td><td style="text-align:center; vertical-align:middle;">51.61</td><td style="text-align:center; vertical-align:middle;">46.67</td><td style="text-align:center; vertical-align:middle;">33.33</td><td style="text-align:center; vertical-align:middle;">28.28</td><td style="text-align:center; vertical-align:middle;">32.48</td><td style="text-align:center; vertical-align:middle;">4.71</td><td style="text-align:center; vertical-align:middle;">30.61</td><td style="text-align:center; vertical-align:middle;">27.31</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>KONI-Llama3.1-8B-R-20250831<br>(KO-REAson-KL3_1-8B-0831; Ours)</b></td>
+      <td style="text-align:center; vertical-align:middle;">8.03</td><td style="text-align:center; vertical-align:middle;">44.64</td><td style="text-align:center; vertical-align:middle;">40.08</td><td style="text-align:center; vertical-align:middle;">37.96</td><td style="text-align:center; vertical-align:middle;">23.33</td><td style="text-align:center; vertical-align:middle;">30.00</td><td style="text-align:center; vertical-align:middle;">38.38</td><td style="text-align:center; vertical-align:middle;">56.39</td><td style="text-align:center; vertical-align:middle;">21.57</td><td style="text-align:center; vertical-align:middle;">30.57</td><td style="text-align:center; vertical-align:middle;">40.13</td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>KONI-7B-R-20250831<br>(KO-REASon-AX3_1-7B-0831; Ours)</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.26</td><td style="text-align:center; vertical-align:middle;">45.57</td><td style="text-align:center; vertical-align:middle;">38.13</td><td style="text-align:center; vertical-align:middle;">52.80</td><td style="text-align:center; vertical-align:middle;">53.33</td><td style="text-align:center; vertical-align:middle;">33.33</td><td style="text-align:center; vertical-align:middle;">36.87</td><td style="text-align:center; vertical-align:middle;"><b>62.86</b></td><td style="text-align:center; vertical-align:middle;">23.43</td><td style="text-align:center; vertical-align:middle;"><u>43.29</u></td><td style="text-align:center; vertical-align:middle;"><u>44.56</u></td>
+    </tr>
+    <tr>
+      <td style="text-align:center; vertical-align:middle;"><b>KO-REASon-7B-Q2_5-0831<br>(Ours)</b></td>
+      <td style="text-align:center; vertical-align:middle;">7.26</td><td style="text-align:center; vertical-align:middle;"><b>46.81</b></td><td style="text-align:center; vertical-align:middle;"><b>44.93</b></td><td style="text-align:center; vertical-align:middle;">48.11</td><td style="text-align:center; vertical-align:middle;">43.33</td><td style="text-align:center; vertical-align:middle;">30.00</td><td style="text-align:center; vertical-align:middle;">42.93</td><td style="text-align:center; vertical-align:middle;">60.65</td><td style="text-align:center; vertical-align:middle;"><b>25.00</b></td><td style="text-align:center; vertical-align:middle;">42.72</td><td style="text-align:center; vertical-align:middle;"><b>45.10</b></td>
+    </tr>
+  </tbody>
+</table>
+---
+## Citation
+The paper will be released soon!
+If you use this model in your work, please cite it as follows:
+```
+@article{KISTI-KONI/KONI-Llama3.1-8B-R-20250831,
+  title={KISTI-KONI/KONI-Llama3.1-8B-R-20250831},
+  author={KISTI, OneLine AI, HAE-RAE and Oracle},
+  year={2025},
+  url={https://huggingface.co/KISTI-KONI/KONI-Llama3.1-8B-R-20250831}
+}
+```
+## Contact
+For any questions contact us via the following email :)
+```
+(KISTI)yangdonghun3@kisti.re.kr
+(OneLineAI/HAE-RAE)spthsrbwls123@yonsei.ac.kr
+```
+## Acknowlegments
+```
+This research is a collaborative project between KISTI, OneLine AI, HAE-RAE and Oracle to investigate open reciepes to build Korean Reasoning Models.<br>
+This research was also supported by the Korea Institute of Science and Technology Information (KISTI) (No.(KISTI) K25L1M1C1), aimed at developing KONI (KISTI Open Neural Intelligence), a large language model specialized in science and technology.
+This work also benefited from the resources and technical support provided by the National Supercomputing Center (KISTI).
+```