Upload README.md with huggingface_hub

README.md

@@ -1,49 +1,179 @@
-Mi:dm 2.0 Base
-
-🤗 Mi:dm 2.0 Models | 📜 Mi:dm 2.0 Technical Report | 📕 Mi:dm 2.0 Technical Blog
-
-
-News 📢
-🔜 (Coming Soon!) GGUF format model files will be available soon for easier local deployment.
-📕2025/08/08: Published a technical blog article about Mi:dm 2.0 Model.
-⚡️2025/07/04: Released Mi:dm 2.0 Model collection on Hugging Face🤗.
-
-Table of Contents
-Overview
-Mi:dm 2.0
-Quickstart
-Evaluation
-Usage
-Run on Friendli.AI
-Run on Your Local Machine
-Deployment
-Tutorials
-More Information
-Limitation
-License
-Contact
-
-
-Overview
-Mi:dm 2.0
-Mi:dm 2.0 is a "Korea-centric AI" model developed using KT's proprietary technology. The term "Korea-centric AI" refers to a model that deeply internalizes the unique values, cognitive frameworks, and commonsense reasoning inherent to Korean society. It goes beyond simply processing or generating Korean text—it reflects a deeper understanding of the socio-cultural norms and values that define Korean society.
+---
+license: mit
+language:
+- en
+- ko
+tags:
+- KT
+- K-intelligence
+- Mi:dm
+inference: true
+pipeline_tag: text-generation
+base_model:
+- K-intelligence/Midm-2.0-Base-Instruct
+
+
+---
+OLLAMA에 추가할 때 Modelfile 참고
+
+```
+FROM ./Midm-2.0-Base-Instruct-f16.gguf
+
+TEMPLATE """<|begin_of_text|><|start_header_id|>system<|end_header_id|>
+Mi:dm(믿:음)은 KT에서 개발한 AI 기반 어시스턴트이다. 너는 Mi:dm으로서 사용자에게 유용하고 안전한 응답을 제공해야 한다.
+Mi:dm은 December 2024까지의 지식으로 학습되었으며 그 외의 지식을 묻는 경우에는 한계를 인정해야 한다.
+어시스턴트는 기본적으로 "한국어"를 사용한다. 사용자의 요청에 따라 생각하고 응답하는 언어는 달라질 수 있으며, 다른 요구사항이 없다면 입력 언어를 따라 응답하라.
+코드 작성 시에는 요구되는 언어의 소스코드로 작성해야 하며, STEM(과학, 기술, 공학, 수학) 분야의 전문 용어는 원문을 그대로 유지하여 출력한다.
+Mi:dm은 사용자 친화적으로 답변을 제공해야 한다. 사용자의 요청이 없다면 기본적으로 경어체를 사용해야 한다.
+사용자의 요청에 따라 유용하고 관련성 있는 답변을 제공해야 한다. 이는 요청의 내용을 반영하여 이루어져야 한다.
+특히, 사용자가 특정 답변의 형태를 요구한 경우 해당 형태를 반드시 준수하여 응답을 제공한다.(예. JSON, SQL, 코드 등) 그렇지 않다면 응답을 제공하기 가장 적절한 형태로 제공한다.
+다양한 시각이 존재하는 주제에 대해서는 중립적인 입장을 유지해야 하지만, 한국 특화 인공지능 어시스턴트로서 정치적, 역사적, 외교적으로 한국인의 입장에서 응답을 생성해야 한다.
+모호하거나 복잡한 요청으로 인해 답변 작성이 어려운 경우, 다시 한 번 생각해보고 추가정보를 요청해야 한다.
+
+Mi:dm은 응답을 제공할 때 어시스턴트의 안전성 측면에서 다음 지침을 *반드시* 준수해야 한다.
+- 비속어와 욕설을 사용하지 않아야 한다.
+- 신뢰할 수 있는 응답을 생성하고, 전문영역에 대한 한계와 불확실성을 인정해야 한다.
+- 사회의 보편적 규범과 가치에 따라 윤리적이고 중립적이어야 하며, 편향성을 지녀서는 안 된다.
+- 인공지능으로서의 정체성을 인지하고 의인화하지 않아야 한다.
+- 개인정보, 사생활 등 민감정보를 포함한 요청에 대한 답변을 거절해야 한다. 다만, 해당정보를 사용할 수 없는 형태(비식별화된 형태)로 제공하는 것은 제한적으로 응답을 허용한다.
+
+이 모든 지침은 응답을 제공할 때 출력되지 않아야 한다.
+
+Mi:dm은 사용자의 요청을 처리하기 위해 제공된 도구(함수)를 호출할 수 있다.
+{{ if .Tools -}}
+Mi:dm은 도구 사용시 아래 규칙을 준수해야 한다.
+- 제공된 도구만 사용하고, 모든 필수 인자를 반드시 포함한다.
+- 주어진 tool_name을 임의로 변경하지 않아야 한다.
+- 도구를 호출하는 경우, 마지막은 도구 호출로 끝내며 그 뒤에 텍스트를 출력하지 않는다.
+- 도구 호출 결과를 활용하여 응답을 생성한다.
+- 도구가 필요하지 않은 경우에는 일반적인 방식으로 응답한다.
+- 도구 호출 정보는 다음과 같이 <tool_call></tool_call> XML 태그 사이에 작성한다.
+<tool_call>{"name": "tool_name", "arguments": {"param":"value"}}</tool_call>
+
+tool_list:[
+    {{- range $i, $tool := .Tools -}}
+        {{- if ne 0 $i }},{{- end -}}
+        {{- $tool -}}
+    {{- end -}}
+]
+{{- end -}}
+{{- if .System -}}
+{{- .System }}
+{{- end -}}
+{{- range $i, $_ := .Messages -}}
+{{- $last := eq (len (slice $.Messages $i)) 1 -}}
+{{- if ne .Role "system" -}}
+<|eot_id|><|start_header_id|>
+{{- .Role -}}
+<|end_header_id|>
+
+{{ if .Content -}}
+{{- .Content -}}
+{{- else if .ToolCalls -}}
+<tool_call>
+{{- range .ToolCalls }}
+{"name": "{{ .Function.Name }}", "parameters": {{ .Function.Arguments }}}
+{{- end }}
+</tool_call>
+{{- end -}}
+{{- if $last -}}
+<|eot_id|><|start_header_id|>assistant<|end_header_id|>
+
+{{ end -}}
+{{- end -}}
+{{- end -}}"""
+
+PARAMETER stop "<|eot_id|>"
+PARAMETER stop "<|end_of_text|>"
+
+LICENSE """MIT License
+
+Copyright (c) 2025 KT Corporation
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE."""
+```
+---
+Thanks to KT
+
+## Mi:dm Official Repo's Description
+
+<p align="center">
+🤗 <a href="https://huggingface.co/collections/K-intelligence/mi-dm-20-6866406c301e5f45a6926af8">Mi:dm 2.0 Models</a> |
+📜 <a href="https://github.com/K-intelligence-Midm/Midm-2.0/blob/main/Mi_dm2_0_technical_report.pdf">Mi:dm 2.0 Technical Report</a> |
+📕 Mi:dm 2.0 Technical Blog*
+</p>
+
+<p align="center"><sub>*To be released soon</sub></p>
+
+<br>
+
+# News 📢
+
+- 🔜 _(Coming Soon!) GGUF format model files will be available soon for easier local deployment._  
+- ⚡️`2025/07/04`: Released Mi:dm 2.0 Model collection on Hugging Face🤗.
+<br>
+<br>
+# Table of Contents
+
+- ___Overview___
+    - [Mi:dm 2.0](#midm-20)
+    - [Quickstart](#quickstart)
+    - [Evaluation](#evaluation)
+- ___Usage___
+    - [Run on Friendli.AI](#run-on-friendliai)
+    - [Run on Your Local Machine](#run-on-your-local-machine)
+    - [Deployment](#deployment)
+    - [Tutorials](#tutorials)
+- ___More Information___
+    - [Limitation](#limitation)
+    - [License](#license)
+    - [Contact](#contact)
+
+<br>
+<br>
+
+# Overview
+
+### Mi:dm 2.0
+
+**Mi:dm 2.0** is a __"Korea-centric AI"__ model developed using KT's proprietary technology. The term __"Korea-centric AI"__ refers to a model that deeply internalizes the unique values, cognitive frameworks, and commonsense reasoning inherent to Korean society. It goes beyond simply processing or generating Korean text—it reflects a deeper understanding of the socio-cultural norms and values that define Korean society.
 
 Mi:dm 2.0 is released in two versions:
 
-Mi:dm 2.0 Base
-An 11.5B parameter dense model designed to balance model size and performance.
-It extends an 8B-scale model by applying the Depth-up Scaling (DuS) method, making it suitable for real-world applications that require both performance and versatility.
+- **Mi:dm 2.0 Base**  
+  An 11.5B parameter dense model designed to balance model size and performance.  
+  It extends an 8B-scale model by applying the Depth-up Scaling (DuS) method, making it suitable for real-world applications that require both performance and versatility.
+
+- **Mi:dm 2.0 Mini**  
+  A lightweight 2.3B parameter dense model optimized for on-device environments and systems with limited GPU resources.  
+  It was derived from the Base model through pruning and distillation to enable compact deployment.
 
-Mi:dm 2.0 Mini
-A lightweight 2.3B parameter dense model optimized for on-device environments and systems with limited GPU resources.
-It was derived from the Base model through pruning and distillation to enable compact deployment.
+> [!Note]
+> Neither the pre-training nor the post-training data includes KT users' data.
 
-Neither the pre-training nor the post-training data includes KT users' data.
+<br>
 
+### Quickstart
 
-Quickstart
 Here is the code snippet to run conversational inference with the model:
 
+```python
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig
 
@@ -82,76 +212,457 @@ output = model.generate(
     do_sample=False,
 )
 print(tokenizer.decode(output[0]))
-
-The transformers library should be version 4.45.0 or higher.
-
-
-Evaluation
-Korean
-Model	Society & Culture	General Knowledge	Instruction Following
-K-Refer*	K-Refer-Hard*	Ko-Sovereign*	HAERAE	Avg.	KMMLU	Ko-Sovereign*	Avg.	Ko-IFEval	Ko-MTBench	Avg.
-Qwen3-4B	53.6	42.9	35.8	50.6	45.7	50.6	42.5	46.5	75.9	63.0	69.4
-Exaone-3.5-2.4B-inst	64.0	67.1	44.4	61.3	59.2	43.5	42.4	43.0	65.4	74.0	68.9
-Mi:dm 2.0-Mini-inst	66.4	61.4	36.7	70.8	58.8	45.1	42.4	43.8	73.3	74.0	73.6
-Qwen3-14B	72.4	65.7	49.8	68.4	64.1	55.4	54.7	55.1	83.6	71	77.3
-Llama-3.1-8B-inst	43.2	36.4	33.8	49.5	40.7	33.0	36.7	34.8	60.1	57	58.5
-Exaone-3.5-7.8B-inst	71.6	69.3	46.9	72.9	65.2	52.6	45.6	49.1	69.1	79.6	74.4
-Mi:dm 2.0-Base-inst	89.6	86.4	56.3	81.5	78.4	57.3	58.0	57.7	82	89.7	85.9
-Model	Comprehension	Reasoning
-K-Prag*	K-Refer-Hard*	Ko-Best	Ko-Sovereign*	Avg.	Ko-Winogrande	Ko-Best	LogicKor	HRM8K	Avg.
-Qwen3-4B	73.9	56.7	91.5	43.5	66.6	67.5	69.2	5.6	56.7	43.8
-Exaone-3.5-2.4B-inst	68.7	58.5	87.2	38.0	62.5	60.3	64.1	7.4	38.5	36.7
-Mi:dm 2.0-Mini-inst	69.5	55.4	80.5	42.5	61.9	61.7	64.5	7.7	39.9	37.4
-Qwen3-14B	86.7	74.0	93.9	52.0	76.8	77.2	75.4	6.4	64.5	48.8
-Llama-3.1-8B-inst	59.9	48.6	77.4	31.5	51.5	40.1	26.0	2.4	30.9	19.8
-Exaone-3.5-7.8B-inst	73.5	61.9	92.0	44.0	67.2	64.6	60.3	8.6	49.7	39.5
-Mi:dm 2.0-Base-inst	86.5	70.8	95.2	53.0	76.1	75.1	73.0	8.6	52.9	44.8
-* indicates KT proprietary evaluation resources.
-
-
-English
-Model	Instruction	Reasoning	Math	Coding	General Knowledge
-IFEval	BBH	GPQA	MuSR	Avg.	GSM8K	MBPP+	MMLU-pro	MMLU	Avg.
-Qwen3-4B	79.7	79.0	39.8	58.5	59.1	90.4	62.4	-	73.3	73.3
-Exaone-3.5-2.4B-inst	81.1	46.4	28.1	49.7	41.4	82.5	59.8	-	59.5	59.5
-Mi:dm 2.0-Mini-inst	73.6	44.5	26.6	51.7	40.9	83.1	60.9	-	56.5	56.5
- 
-Qwen3-14B	83.9	83.4	49.8	57.7	63.6	88.0	73.4	70.5	82.7	76.6
-Llama-3.1-8B-inst	79.9	60.3	21.6	50.3	44.1	81.2	81.8	47.6	70.7	59.2
-Exaone-3.5-7.8B-inst	83.6	50.1	33.1	51.2	44.8	81.1	79.4	40.7	69.0	54.8
-Mi:dm 2.0-Base-inst	84.0	77.7	33.5	51.9	54.4	91.6	77.5	53.3	73.7	63.5
-
-Usage
-Run on Friendli.AI
-You can try our model immediately via Friendli.AI. Simply click Deploy and then Friendli Endpoints.
-
-Please note that a login to Friendli.AI is required after your fifth chat interaction.
-
-Left Image Right Image
-
-Run on Your Local Machine
-We provide a detailed description about running Mi:dm 2.0 on your local machine using llama.cpp, LM Studio, and Ollama. Please check our github for more information
-
-Deployment
-To serve Mi:dm 2.0 using vLLM(>=0.8.0) with an OpenAI-compatible API:
-
+```
+
+> [!NOTE]
+> The `transformers` library should be version `4.45.0` or higher.
+
+<br>
+
+# Evaluation
+
+
+#### Korean
+
+<!-- first half table-->
+<table>
+<tr>
+  <th rowspan="2">Model</th>
+  <th colspan="5" align="center">Society & Culture</th>
+  <th colspan="3" align="center">General Knowledge</th>
+  <th colspan="3" align="center">Instruction Following</th>
+</tr>
+<tr>
+  <th align="center">K-Refer<sup>*</sup></th>
+  <th align="center">K-Refer-Hard<sup>*</sup></th>
+  <th align="center">Ko-Sovereign<sup>*</sup></th>
+  <th align="center">HAERAE</th>
+  <th align="center">Avg.</th>
+  <th align="center">KMMLU</th>
+  <th align="center">Ko-Sovereign<sup>*</sup></th>
+  <th align="center">Avg.</th>
+  <th align="center">Ko-IFEval</th>
+  <th align="center">Ko-MTBench</th>
+  <th align="center">Avg.</th>
+</tr>
+
+<!-- Small Models -->
+<tr>
+  <td><strong>Qwen3-4B</strong></td>
+  <td align="center">53.6</td>
+  <td align="center">42.9</td>
+  <td align="center">35.8</td>
+  <td align="center">50.6</td>
+  <td align="center">45.7</td>
+  <td align="center"><strong>50.6</strong></td>
+  <td align="center"><strong>42.5</strong></td>
+  <td align="center"><strong>46.5</strong></td>
+  <td align="center"><strong>75.9</strong></td>
+  <td align="center">63.0</td>
+  <td align="center">69.4</td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-2.4B-inst</strong></td>
+  <td align="center">64.0</td>
+  <td align="center"><strong>67.1</strong></td>
+  <td align="center"><strong>44.4</strong></td>
+  <td align="center">61.3</td>
+  <td align="center"><strong>59.2</strong></td>
+  <td align="center">43.5</td>
+  <td align="center">42.4</td>
+  <td align="center">43.0</td>
+  <td align="center">65.4</td>
+  <td align="center"><strong>74.0</strong></td>
+  <td align="center">68.9</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Mini-inst</strong></td>
+  <td align="center"><strong>66.4</strong></td>
+  <td align="center">61.4</td>
+  <td align="center">36.7</td>
+  <td align="center"><strong>70.8</strong></td>
+  <td align="center">58.8</td>
+  <td align="center">45.1</td>
+  <td align="center">42.4</td>
+  <td align="center">43.8</td>
+  <td align="center">73.3</td>
+  <td align="center"><strong>74.0</strong></td>
+  <td align="center"><strong>73.6</strong></td>
+</tr>
+
+<!-- Spacer row -->
+<tr><td colspan="13"> </td></tr>
+
+<!-- Large Models -->
+<tr>
+  <td><strong>Qwen3-14B</strong></td>
+  <td align="center">72.4</td>
+  <td align="center">65.7</td>
+  <td align="center">49.8</td>
+  <td align="center">68.4</td>
+  <td align="center">64.1</td>
+  <td align="center">55.4</td>
+  <td align="center">54.7</td>
+  <td align="center">55.1</td>
+  <td align="center"><strong>83.6</strong></td>
+  <td align="center">71</td>
+  <td align="center">77.3</td>
+</tr>
+<tr>
+  <td><strong>Llama-3.1-8B-inst</strong></td>
+  <td align="center">43.2</td>
+  <td align="center">36.4</td>
+  <td align="center">33.8</td>
+  <td align="center">49.5</td>
+  <td align="center">40.7</td>
+  <td align="center">33.0</td>
+  <td align="center">36.7</td>
+  <td align="center">34.8</td>
+  <td align="center">60.1</td>
+  <td align="center">57</td>
+  <td align="center">58.5</td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-7.8B-inst</strong></td>
+  <td align="center">71.6</td>
+  <td align="center">69.3</td>
+  <td align="center">46.9</td>
+  <td align="center">72.9</td>
+  <td align="center">65.2</td>
+  <td align="center">52.6</td>
+  <td align="center">45.6</td>
+  <td align="center">49.1</td>
+  <td align="center">69.1</td>
+  <td align="center">79.6</td>
+  <td align="center">74.4</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Base-inst</strong></td>
+  <td align="center"><strong>89.6</strong></td>
+  <td align="center"><strong>86.4</strong></td>
+  <td align="center"><strong>56.3</strong></td>
+  <td align="center"><strong>81.5</strong></td>
+  <td align="center"><strong>78.4</strong></td>
+  <td align="center"><strong>57.3</strong></td>
+  <td align="center"><strong>58.0</strong></td>
+  <td align="center"><strong>57.7</strong></td>
+  <td align="center">82</td>
+  <td align="center"><strong>89.7</strong></td>
+  <td align="center"><strong>85.9</strong></td>
+</tr>
+</table>
+
+<!-- second half table-->
+<table>
+<tr>
+  <th rowspan="2" align="center">Model</th>
+  <th colspan="5" align="center">Comprehension</th>
+  <th colspan="5" align="center">Reasoning</th>
+</tr>
+<tr>
+  <th align="center">K-Prag<sup>*</sup></th>
+  <th align="center">K-Refer-Hard<sup>*</sup></th>
+  <th align="center">Ko-Best</th>
+  <th align="center">Ko-Sovereign<sup>*</sup></th>
+  <th align="center">Avg.</th>
+  <th align="center">Ko-Winogrande</th>
+  <th align="center">Ko-Best</th>
+  <th align="center">LogicKor</th>
+  <th align="center">HRM8K</th>
+  <th align="center">Avg.</th>
+</tr>
+
+<!-- Small Models -->
+<tr>
+  <td><strong>Qwen3-4B</strong></td>
+  <td align="center"><strong>73.9<strong></td>
+  <td align="center">56.7</td>
+  <td align="center"><strong>91.5</strong></td>
+  <td align="center"><strong>43.5</strong></td>
+  <td align="center"><strong>66.6</strong></td>
+  <td align="center"><strong>67.5</strong></td>
+  <td align="center"><strong>69.2</strong></td>
+  <td align="center">5.6</td>
+  <td align="center"><strong>56.7</strong></td>
+  <td align="center"><strong>43.8</strong></td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-2.4B-inst</strong></td>
+  <td align="center">68.7</td>
+  <td align="center"><strong>58.5</strong></td>
+  <td align="center">87.2</td>
+  <td align="center">38.0</td>
+  <td align="center">62.5</td>
+  <td align="center">60.3</td>
+  <td align="center">64.1</td>
+  <td align="center">7.4</td>
+  <td align="center">38.5</td>
+  <td align="center">36.7</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Mini-inst</strong></td>
+  <td align="center">69.5</td>
+  <td align="center">55.4</td>
+  <td align="center">80.5</td>
+  <td align="center">42.5</td>
+  <td align="center">61.9</td>
+  <td align="center">61.7</td>
+  <td align="center">64.5</td>
+  <td align="center"><strong>7.7</strong></td>
+  <td align="center">39.9</td>
+  <td align="center">37.4</td>
+</tr>
+
+<!-- Visual Spacer -->
+<tr><td colspan="11"> </td></tr>
+
+<!-- Large Models -->
+<tr>
+  <td><strong>Qwen3-14B</strong></td>
+  <td align="center"><strong>86.7</strong></td>
+  <td align="center"><strong>74.0</strong></td>
+  <td align="center">93.9</td>
+  <td align="center">52.0</td>
+  <td align="center"><strong>76.8</strong></td>
+  <td align="center"><strong>77.2</strong></td>
+  <td align="center"><strong>75.4</strong></td>
+  <td align="center">6.4</td>
+  <td align="center"><strong>64.5</strong></td>
+  <td align="center"><strong>48.8</strong></td>
+</tr>
+<tr>
+  <td><strong>Llama-3.1-8B-inst</strong></td>
+  <td align="center">59.9</td>
+  <td align="center">48.6</td>
+  <td align="center">77.4</td>
+  <td align="center">31.5</td>
+  <td align="center">51.5</td>
+  <td align="center">40.1</td>
+  <td align="center">26.0</td>
+  <td align="center">2.4</td>
+  <td align="center">30.9</td>
+  <td align="center">19.8</td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-7.8B-inst</strong></td>
+  <td align="center">73.5</td>
+  <td align="center">61.9</td>
+  <td align="center">92.0</td>
+  <td align="center">44.0</td>
+  <td align="center">67.2</td>
+  <td align="center">64.6</td>
+  <td align="center">60.3</td>
+  <td align="center"><strong>8.6</strong></td>
+  <td align="center">49.7</td>
+  <td align="center">39.5</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Base-inst</strong></td>
+  <td align="center">86.5</td>
+  <td align="center">70.8</td>
+  <td align="center"><strong>95.2</strong></td>
+  <td align="center"><strong>53.0</strong></td>
+  <td align="center">76.1</td>
+  <td align="center">75.1</td>
+  <td align="center">73.0</td>
+  <td align="center"><strong>8.6</strong></td>
+  <td align="center">52.9</td>
+  <td align="center">44.8</td>
+</tr>
+</table>
+
+`*` indicates KT proprietary evaluation resources.
+
+<br>
+
+
+#### English
+
+
+<table>
+<tr>
+  <th rowspan="2" align="center">Model</th>
+  <th align="center">Instruction</th>
+  <th colspan="4" align="center">Reasoning</th>
+  <th align="center">Math</th>
+  <th align="center">Coding</th>
+  <th colspan="3" align="center">General Knowledge</th>
+</tr>
+<tr>
+  <th align="center">IFEval</th>
+  <th align="center">BBH</th>
+  <th align="center">GPQA</th>
+  <th align="center">MuSR</th>
+  <th align="center">Avg.</th>
+  <th align="center">GSM8K</th>
+  <th align="center">MBPP+</th>
+  <th align="center">MMLU-pro</th>
+  <th align="center">MMLU</th>
+  <th align="center">Avg.</th>
+</tr>
+
+<!-- Small Models -->
+<tr>
+  <td><strong>Qwen3-4B</strong></td>
+  <td align="center">79.7</td>
+  <td align="center"><strong>79.0</strong></td>
+  <td align="center"><strong>39.8</strong></td>
+  <td align="center"><strong>58.5</strong></td>
+  <td align="center"><strong>59.1</strong></td>
+  <td align="center"><strong>90.4</strong></td>
+  <td align="center">62.4</td>
+  <td align="center">-</td>
+  <td align="center"><strong>73.3</strong></td>
+  <td align="center"><strong>73.3</strong></td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-2.4B-inst</strong></td>
+  <td align="center"><strong>81.1</strong></td>
+  <td align="center">46.4</td>
+  <td align="center">28.1</td>
+  <td align="center">49.7</td>
+  <td align="center">41.4</td>
+  <td align="center">82.5</td>
+  <td align="center">59.8</td>
+  <td align="center">-</td>
+  <td align="center">59.5</td>
+  <td align="center">59.5</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Mini-inst</strong></td>
+  <td align="center">73.6</td>
+  <td align="center">44.5</td>
+  <td align="center">26.6</td>
+  <td align="center">51.7</td>
+  <td align="center">40.9</td>
+  <td align="center">83.1</td>
+  <td align="center"><strong>60.9</strong></td>
+  <td align="center">-</td>
+  <td align="center">56.5</td>
+  <td align="center">56.5</td>
+</tr>
+
+<tr><td colspan="11">&nbsp;</td></tr>
+
+<!-- Large Models -->
+<tr>
+  <td><strong>Qwen3-14B</strong></td>
+  <td align="center">83.9</td>
+  <td align="center"><strong>83.4</strong></td>
+  <td align="center"><strong>49.8</strong></td>
+  <td align="center"><strong>57.7</strong></td>
+  <td align="center"><strong>63.6</strong></td>
+  <td align="center">88.0</td>
+  <td align="center">73.4</td>
+  <td align="center"><strong>70.5</strong></td>
+  <td align="center"><strong>82.7</strong></td>
+  <td align="center"><strong>76.6</strong></td>
+</tr>
+<tr>
+  <td><strong>Llama-3.1-8B-inst</strong></td>
+  <td align="center">79.9</td>
+  <td align="center">60.3</td>
+  <td align="center">21.6</td>
+  <td align="center">50.3</td>
+  <td align="center">44.1</td>
+  <td align="center">81.2</td>
+  <td align="center"><strong>81.8</strong></td>
+  <td align="center">47.6</td>
+  <td align="center">70.7</td>
+  <td align="center">59.2</td>
+</tr>
+<tr>
+  <td><strong>Exaone-3.5-7.8B-inst</strong></td>
+  <td align="center">83.6</td>
+  <td align="center">50.1</td>
+  <td align="center">33.1</td>
+  <td align="center">51.2</td>
+  <td align="center">44.8</td>
+  <td align="center">81.1</td>
+  <td align="center">79.4</td>
+  <td align="center">40.7</td>
+  <td align="center">69.0</td>
+  <td align="center">54.8</td>
+</tr>
+<tr>
+  <td><strong>Mi:dm 2.0-Base-inst</strong></td>
+  <td align="center"><strong>84.0</strong></td>
+  <td align="center">77.7</td>
+  <td align="center">33.5</td>
+  <td align="center">51.9</td>
+  <td align="center">54.4</td>
+  <td align="center"><strong>91.6</strong></td>
+  <td align="center">77.5</td>
+  <td align="center">53.3</td>
+  <td align="center">73.7</td>
+  <td align="center">63.5</td>
+</tr>
+</table>
+
+
+<br>
+
+# Usage
+
+### Run on Friendli.AI
+You can try our model immediately via `Friendli.AI`. Simply click `Deploy` and then `Friendli Endpoints`. 
+
+> [!Note]
+> Please note that a login to `Friendli.AI` is required after your fifth chat interaction.
+
+<p>
+  <img src="./assets/image_1.png" alt="Left Image" width="36%" style="display:inline-block; margin-right:2%">
+  <img src="./assets/image_2.png" alt="Right Image" width="36%" style="display:inline-block">
+</p>
+
+
+### Run on Your Local Machine
+We provide a detailed description about running Mi:dm 2.0 on your local machine using llama.cpp, LM Studio, and Ollama. Please check our [github](https://github.com/K-intelligence-Midm/Midm-2.0) for more information
+
+
+### Deployment
+
+To serve Mi:dm 2.0 using [vLLM](https://github.com/vllm-project/vllm)(`>=0.8.0`) with an OpenAI-compatible API:
+```bash
 vllm serve K-intelligence/Midm-2.0-Base-Instruct
+```
+
 
-Tutorials
-To help our end-users easily use Mi:dm 2.0, we have provided comprehensive tutorials on github.
+### Tutorials
+To help our end-users easily use Mi:dm 2.0, we have provided comprehensive tutorials on [github](https://github.com/K-intelligence-Midm/Midm-2.0). 
+<br>
 
+<br>
+<br>
 
+# More Information
 
-More Information
-Limitation
-The training data for both Mi:dm 2.0 models consists primarily of English and Korean. Understanding and generation in other languages are not guaranteed.
+### Limitation
+* The training data for both Mi:dm 2.0 models consists primarily of English and Korean. Understanding and generation in other languages are not guaranteed.
+  
+* The model is not guaranteed to provide reliable advice in fields that require professional expertise, such as law, medicine, or finance.
 
-The model is not guaranteed to provide reliable advice in fields that require professional expertise, such as law, medicine, or finance.
+* Researchers have made efforts to exclude unethical content from the training data — such as profanity, slurs, bias, and discriminatory language. However, despite these efforts, the model may still produce inappropriate expressions or factual inaccuracies.
 
-Researchers have made efforts to exclude unethical content from the training data — such as profanity, slurs, bias, and discriminatory language. However, despite these efforts, the model may still produce inappropriate expressions or factual inaccuracies.
 
-License
-Mi:dm 2.0 is licensed under the MIT License.
+### License
 
-Contact
-Mi:dm 2.0 Technical Inquiries: midm-llm@kt.com
+Mi:dm 2.0 is licensed under the [MIT License](./LICENSE).
+ 
+<!-- ### Citation
+ 
+```
+@misc{,
+      title={}, 
+      author={},
+      year={2025},
+      eprint={},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={}, 
+}
+``` -->
+### Contact 
+Mi:dm 2.0 Technical Inquiries: midm-llm@kt.com
+
+<br>