初始化项目，由ModelHub XC社区提供模型

Model: Mungert/Midm-2.0-Base-Instruct-GGUF Source: Original Platform
2026-10-18 04:26:17 +08:00
commit d467883ab2
34 changed files with 852 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,81 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-f16_q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-bf16_q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-f16_q6_k.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-bf16_q6_k.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-f16_q4_k.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-bf16_q4_k.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q2_k_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q3_k_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_k_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_k_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q6_k_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q2_k_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q2_k_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q3_k_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q3_k_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_k_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_k_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_k_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_k_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q6_k_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_1.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_0_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q4_1_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_1.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_0_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-q5_1_l.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq1_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq1_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq2_xs.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq2_xxs.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq2_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq2_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq3_xs.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq3_xxs.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq3_s.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq3_m.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq4_xs.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-iq4_nl.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-tq1_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-tq2_0.gguf filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text
+Midm-2.0-Base-Instruct-bf16.gguf filter=lfs diff=lfs merge=lfs -text
--- a/Midm-2.0-Base-Instruct-bf16.gguf
+++ b/Midm-2.0-Base-Instruct-bf16.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8af30758ce47b8592a2de551094ba20e5f0afbd8fde1ce371aaf5ed0e92e1021
+size 23097821216
--- a/Midm-2.0-Base-Instruct-bf16_q8_0.gguf
+++ b/Midm-2.0-Base-Instruct-bf16_q8_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e76dcba1fa165eec0c6df620e736d450c39be01c07b3003765e3104ef238b4ec
+size 16869279776
--- a/Midm-2.0-Base-Instruct-f16_q8_0.gguf
+++ b/Midm-2.0-Base-Instruct-f16_q8_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49da267c27e331cc69bea9c6d501aadad0fdbb17f09a45af534bc8f49a8a7527
+size 16869279776
--- a/Midm-2.0-Base-Instruct-iq1_m.gguf
+++ b/Midm-2.0-Base-Instruct-iq1_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3394283ee7bc92b32cdca5c8317e32a01780c31433d8d76425f5793c564640dc
+size 3701447008
--- a/Midm-2.0-Base-Instruct-iq1_s.gguf
+++ b/Midm-2.0-Base-Instruct-iq1_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a832dc102751fa04c71b87f1c335deb95ac4cf3098ac4820cfd41838b5e16aa2
+size 3445725536
--- a/Midm-2.0-Base-Instruct-iq2_m.gguf
+++ b/Midm-2.0-Base-Instruct-iq2_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb689c2af6e2d1253e3d5d538bfa198ad313bf60d3a3d51663af61edcaac6d10
+size 4417624416
--- a/Midm-2.0-Base-Instruct-iq2_s.gguf
+++ b/Midm-2.0-Base-Instruct-iq2_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e85975eddccd813bfe66f6ed3eb87dea5b6fefca9884a8649164a0c6e2addbb9
+size 4223113568
--- a/Midm-2.0-Base-Instruct-iq2_xs.gguf
+++ b/Midm-2.0-Base-Instruct-iq2_xs.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af508853ad98d1c844354000e069c61e8243e13175388cb7ec8002da4678de6e
+size 4097808736
--- a/Midm-2.0-Base-Instruct-iq2_xxs.gguf
+++ b/Midm-2.0-Base-Instruct-iq2_xxs.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c40dcd890bdbf6eaf9261e917d6900988ac525542219028a46bb159736050d66
+size 3798047072
--- a/Midm-2.0-Base-Instruct-iq3_m.gguf
+++ b/Midm-2.0-Base-Instruct-iq3_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85b079c2539012999253d476066740abf2319baf0bac545c04638c410f17304b
+size 5589014880
--- a/Midm-2.0-Base-Instruct-iq3_s.gguf
+++ b/Midm-2.0-Base-Instruct-iq3_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0355cbabd0107afc37ee837a42e3e415231adec4b4bada11ca14d1b993fda93c
+size 5589014880
--- a/Midm-2.0-Base-Instruct-iq3_xs.gguf
+++ b/Midm-2.0-Base-Instruct-iq3_xs.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d50574dde73e3a5aeb76b5a9b93e6b3b066b3755503bc49b55efba4cc7a97519
+size 5064137056
--- a/Midm-2.0-Base-Instruct-iq3_xxs.gguf
+++ b/Midm-2.0-Base-Instruct-iq3_xxs.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6b055cc3deb47cb479fcf7329153a7c70e868e70f64615f2fbb5ebfd9e0be9f
+size 4935162208
--- a/Midm-2.0-Base-Instruct-iq4_nl.gguf
+++ b/Midm-2.0-Base-Instruct-iq4_nl.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6679a6839172d626fe681249fd292d8fc06e573d49a5af4da0f1f94e72a72bfb
+size 6665388128
--- a/Midm-2.0-Base-Instruct-iq4_xs.gguf
+++ b/Midm-2.0-Base-Instruct-iq4_xs.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3785ac6b7b13780ae0f682f8a6d9d7bedffa73a3a705dce850048f705e4fd138
+size 6327706720
--- a/Midm-2.0-Base-Instruct-q2_k_m.gguf
+++ b/Midm-2.0-Base-Instruct-q2_k_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a08608e7f4165ab6c338efaa24edc433d97e8d2e2102b294838c2e1497dbd7e3
+size 4683908960
--- a/Midm-2.0-Base-Instruct-q2_k_s.gguf
+++ b/Midm-2.0-Base-Instruct-q2_k_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a06e553512dcd33ea0d4b9f204c9429f99c975ff5dc0fe1b4721d0c64a6ef42f
+size 4540963168
--- a/Midm-2.0-Base-Instruct-q3_k_m.gguf
+++ b/Midm-2.0-Base-Instruct-q3_k_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:686edae9970774a0c2e7b735228dd7eea66ab17b9f8970fc3303e18aba7453c5
+size 5876664160
--- a/Midm-2.0-Base-Instruct-q3_k_s.gguf
+++ b/Midm-2.0-Base-Instruct-q3_k_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:728002cfe779fa2709e8f5b3303ffed9a5d74e84580e7ec5b0aa4180ede017b4
+size 5733718368
--- a/Midm-2.0-Base-Instruct-q4_0.gguf
+++ b/Midm-2.0-Base-Instruct-q4_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a4d5002a9cb0d74db4e1a2b756ce91367a8782a5189b4a6285284f6f5750cdf
+size 6501480800
--- a/Midm-2.0-Base-Instruct-q4_1.gguf
+++ b/Midm-2.0-Base-Instruct-q4_1.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:054fbfc34d61d28eda81151ac2c88f1683fe00f942d604c65f8282a12385096b
+size 7223060832
--- a/Midm-2.0-Base-Instruct-q4_k_m.gguf
+++ b/Midm-2.0-Base-Instruct-q4_k_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14eb38af9d8a18041ad93430b71b1e49e30562796c2d9e617c7cac6889ec85de
+size 7105660768
--- a/Midm-2.0-Base-Instruct-q4_k_s.gguf
+++ b/Midm-2.0-Base-Instruct-q4_k_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11a5a39cc7d149c2fcd125c33b45abb3e5f1085e5843a59a630b7d06555eba5d
+size 6885918560
--- a/Midm-2.0-Base-Instruct-q5_0.gguf
+++ b/Midm-2.0-Base-Instruct-q5_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a034810cdfad549686556cf9bb7f67d0780e73c1ca5a083614fb937ee61d998b
+size 7944640864
--- a/Midm-2.0-Base-Instruct-q5_1.gguf
+++ b/Midm-2.0-Base-Instruct-q5_1.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b2454e9cc0e22ab1880967bbf861df7f64a4e0a97181f55adfda98d2328e0bc
+size 8666220896
--- a/Midm-2.0-Base-Instruct-q5_k_m.gguf
+++ b/Midm-2.0-Base-Instruct-q5_k_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa1cf1ab9a8f84b6e643578e071f660b2cde3ee88ad82e9ae2f1be0c47e0c863
+size 8310409056
--- a/Midm-2.0-Base-Instruct-q5_k_s.gguf
+++ b/Midm-2.0-Base-Instruct-q5_k_s.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:776309e5da48b7647ccf738fd19ebe08c579c11794e7239a38f2e98b11262505
+size 8193427296
--- a/Midm-2.0-Base-Instruct-q6_k_m.gguf
+++ b/Midm-2.0-Base-Instruct-q6_k_m.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:035382feff85503084293534923887f811300de079a19e00cfe098b7f07e1a27
+size 9477998432
--- a/Midm-2.0-Base-Instruct-q8_0.gguf
+++ b/Midm-2.0-Base-Instruct-q8_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9d0cd0b59ceab4bc3557b485f14b247658f8663edf395d5fcd34d102a614a5f
+size 12274120736
--- a/Midm-2.0-Base-Instruct-tq1_0.gguf
+++ b/Midm-2.0-Base-Instruct-tq1_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef7855ba35d9566fae1faefaaabb735e5544307af0dee550f735fdb1be8e1052
+size 2959720544
--- a/Midm-2.0-Base-Instruct-tq2_0.gguf
+++ b/Midm-2.0-Base-Instruct-tq2_0.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d280c434e9394de8133b950cc07f9cee633cfccd48edac4009df7aea497b4d70
+size 3450454112
--- a/Midm-2.0-Base-Instruct.imatrix
+++ b/Midm-2.0-Base-Instruct.imatrix
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f64aebaa6f81304631dda4df6b32b45f57e6f30d0cb584bdf266309aff0028a3
+size 7482300
--- a/README.md
+++ b/README.md
@@ -0,0 +1,675 @@
+---
+license: mit
+language:
+- en
+- ko
+tags:
+- KT
+- K-intelligence
+- Mi:dm
+inference: true
+pipeline_tag: text-generation
+library_name: transformers
+---
+
+# <span style="color: #7FFF7F;">Midm-2.0-Base-Instruct GGUF Models</span>
+
+
+## <span style="color: #7F7FFF;">Model Generation Details</span>
+
+This model was generated using [llama.cpp](https://github.com/ggerganov/llama.cpp) at commit [`21c02174`](https://github.com/ggerganov/llama.cpp/commit/21c021745d781edf9c44b4972ef6cbbf53b0ecff).
+
+
+
+
+
+---
+
+## <span style="color: #7FFF7F;">Quantization Beyond the IMatrix</span>
+
+I've been experimenting with a new quantization approach that selectively elevates the precision of key layers beyond what the default IMatrix configuration provides.
+
+In my testing, standard IMatrix quantization underperforms at lower bit depths, especially with Mixture of Experts (MoE) models. To address this, I'm using the `--tensor-type` option in `llama.cpp` to manually "bump" important layers to higher precision. You can see the implementation here:  
+👉 [Layer bumping with llama.cpp](https://github.com/Mungert69/GGUFModelBuilder/blob/main/model-converter/tensor_list_builder.py)
+
+While this does increase model file size, it significantly improves precision for a given quantization level.
+
+### **I'd love your feedback—have you tried this? How does it perform for you?**
+
+
+
+
+---
+
+<a href="https://readyforquantum.com/huggingface_gguf_selection_guide.html" style="color: #7FFF7F;">
+  Click here to get info on choosing the right GGUF model format
+</a>
+
+---
+
+
+
+<!--Begin Original Model Card-->
+
+
+
+<p align="center">
+<br>
+    <span style="font-size: 60px; font-weight: bold;">Mi:dm 2.0 Base</span>
+</br>
+</p>
+
+<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 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.
+
+<br>
+
+### Quickstart
+
+Here is the code snippet to run conversational inference with the model:
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig
+
+model_name = "K-intelligence/Midm-2.0-Base-Instruct"
+
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.bfloat16,
+    trust_remote_code=True,
+    device_map="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+generation_config = GenerationConfig.from_pretrained(model_name)
+
+prompt = "KT에 대해 소개해줘"
+
+# message for inference
+messages = [
+    {"role": "system", 
+     "content": "Mi:dm(믿:음)은 KT에서 개발한 AI 기반 어시스턴트이다."},
+    {"role": "user", "content": prompt}
+]
+
+input_ids = tokenizer.apply_chat_template(
+    messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_tensors="pt"
+)
+
+output = model.generate(
+    input_ids.to("cuda"),
+    generation_config=generation_config,
+    eos_token_id=tokenizer.eos_token_id,
+    max_new_tokens=128,
+    do_sample=False,
+)
+print(tokenizer.decode(output[0]))
+```
+
+> [!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](https://github.com/K-intelligence-Midm/Midm-2.0). 
+<br>
+
+<br>
+<br>
+
+# 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.
+  
+* 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.
+
+
+### License
+
+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>
+
+
+
+<!--End Original Model Card-->
+
+---
+
+# <span id="testllm" style="color: #7F7FFF;">🚀 If you find these models useful</span>
+
+Help me test my **AI-Powered Quantum Network Monitor Assistant** with **quantum-ready security checks**:  
+
+👉 [Quantum Network Monitor](https://readyforquantum.com/?assistant=open&utm_source=huggingface&utm_medium=referral&utm_campaign=huggingface_repo_readme)  
+
+
+The full Open Source Code for the Quantum Network Monitor Service available at my github repos ( repos with NetworkMonitor in the name) : [Source Code Quantum Network Monitor](https://github.com/Mungert69). You will also find the code I use to quantize the models if you want to do it yourself [GGUFModelBuilder](https://github.com/Mungert69/GGUFModelBuilder)
+
+💬 **How to test**:  
+ Choose an **AI assistant type**:  
+   - `TurboLLM` (GPT-4.1-mini)  
+   - `HugLLM` (Hugginface Open-source models)  
+   - `TestLLM` (Experimental CPU-only)  
+
+### **What I’m Testing**  
+I’m pushing the limits of **small open-source models for AI network monitoring**, specifically:  
+- **Function calling** against live network services  
+- **How small can a model go** while still handling:  
+  - Automated **Nmap security scans**  
+  - **Quantum-readiness checks**  
+  - **Network Monitoring tasks**  
+
+🟡 **TestLLM** – Current experimental model (llama.cpp on 2 CPU threads on huggingface docker space):  
+- ✅ **Zero-configuration setup**  
+- ⏳ 30s load time (slow inference but **no API costs**) . No token limited as the cost is low.
+- 🔧 **Help wanted!** If you’re into **edge-device AI**, let’s collaborate!  
+
+### **Other Assistants**  
+🟢 **TurboLLM** – Uses **gpt-4.1-mini** :
+- **It performs very well but unfortunatly OpenAI charges per token. For this reason tokens usage is limited. 
+- **Create custom cmd processors to run .net code on Quantum Network Monitor Agents**
+- **Real-time network diagnostics and monitoring**
+- **Security Audits**
+- **Penetration testing** (Nmap/Metasploit)  
+
+🔵 **HugLLM** – Latest Open-source models:  
+- 🌐 Runs on Hugging Face Inference API. Performs pretty well using the lastest models hosted on Novita.
+
+### 💡 **Example commands you could test**:  
+1. `"Give me info on my websites SSL certificate"`  
+2. `"Check if my server is using quantum safe encyption for communication"`  
+3. `"Run a comprehensive security audit on my server"`
+4. '"Create a cmd processor to .. (what ever you want)" Note you need to install a [Quantum Network Monitor Agent](https://readyforquantum.com/Download/?utm_source=huggingface&utm_medium=referral&utm_campaign=huggingface_repo_readme) to run the .net code on. This is a very flexible and powerful feature. Use with caution!
+
+### Final Word
+
+I fund the servers used to create these model files, run the Quantum Network Monitor service, and pay for inference from Novita and OpenAI—all out of my own pocket. All the code behind the model creation and the Quantum Network Monitor project is [open source](https://github.com/Mungert69). Feel free to use whatever you find helpful.
+
+If you appreciate the work, please consider [buying me a coffee](https://www.buymeacoffee.com/mahadeva) ☕. Your support helps cover service costs and allows me to raise token limits for everyone.
+
+I'm also open to job opportunities or sponsorship.
+
+Thank you! 😊