---
library_name: transformers

---

# HumanF-MarkrAI/Gukbap-Gemma3-4B-VL🍚

## Model Details🍚

### Model Description
- **Developed by:** HumanF-MarkrAI
- **Model type:** Korean-VL-Gemma3-4B
- **Language(s):** Korean + English
- **Context Length:** 2048
- **License:** cc-by-4.0
- **Finetuned from model:** [google/gemma-3-4b-it](https://huggingface.co/google/gemma-3-4b-it).  
  
### Model Sources
When training, we used `H100 80GB GPU`x4.

  
### Implications🍚
If you want to know our model's details, please see [🔥Gukbap-LMM Blog🔥](https://kyujinpy.tistory.com/169).  

  
### Training Method (SFT)🧐
The following papers contain the foundational methodologies for the dataset and training methods we are currently proceeding.  
- [LIMA](https://arxiv.org/abs/2305.11206).
- [Gemma3](https://arxiv.org/abs/2503.19786).  

  
### SFT Text-Datasets (Private)
When we made the `Open-Source based dataset`, we use `microsoft/WizardLM-2-8x22B` through [DeepInfra](https://deepinfra.com/).  
Our datasets are made by `Evolving system`, which is propsed by [WizardLM](https://wizardlm.github.io/WizardLM2/).
In training, we used 1849 training dataset, and 200 validation dataset.
  
- **Wizard-Korea-Datasets:** [MarkrAI/Markr_WizardLM_train_ver4](https://huggingface.co/datasets/MarkrAI/Markr_WizardLM_train_ver4).   
> Learning rate: 1e-5; Epoch: 5

    
## Benchmakrs🤗
   
### Korean MM Benchmark Score (Zero-shot)

We internally evaluated [🔥our code🔥](https://github.com/Marker-Inc-Korea/KoVLMEval).  
We utilized **gpt-4o-2024-08-06** in `K-LLAVA-W` evaluation.  
  
| Model | K-MMBench | K-MMStar | K-DTCBench | K-LLAVA-W | AVG |
|:---------:|:-----:|:------:|:-----:|:-----:|:----:|
| **Gukbap-Gemma3-4B**🍚 | 74.73 | 40.67 | 44.17 | **60.00** | **54.89** |
| gemma-3-4b-it | 75.84 | 40.60 | 49.58 | **62.67** | 57.17 |
| Gukbap-Gemma2-9B-VL🍚 | 80.16 | 54.20 | 52.92 | 63.83 | 62.78 |
| Ovis1.6-Gemma2-9B | 52.46 | 50.40 | 47.08 | 55.67 | 51.40 |
| VARCO-VISION-14B | **87.16** | **58.13** | **85.42** | 51.17 | **70.47** |
| llama-3.2-Korean-Bllossom-AICA-5B | 26.01 | 21.60 | 17.08 | 45.33 | 27.51 |
   
### MM Benchmarks
- Global MM Bench dataset: [OpenCampass MM leaderboard](https://rank.opencompass.org.cn/leaderboard-multimodal)
- Korean MM Bench dataset: [NCSOFT](https://huggingface.co/NCSOFT).

     
## Gukbap-VL Series models🍚🍚
- [HumanF-MarkrAI/Gukbap-Gemma2-9B-VL](https://huggingface.co/HumanF-MarkrAI/Gukbap-Gemma2-9B-VL)
- [HumanF-MarkrAI/Gukbap-Gemma3-4B-VL](https://huggingface.co/HumanF-MarkrAI/Gukbap-Gemma3-4B-VL)
- [HumanF-MarkrAI/Gukbap-Gemma3-12B-VL](https://huggingface.co/HumanF-MarkrAI/Gukbap-Gemma3-12B-VL)
- [HumanF-MarkrAI/Gukbap-Gemma3-27B-VL](https://huggingface.co/HumanF-MarkrAI/Gukbap-Gemma3-27B-VL)
   
  
## BibTeX
```
@article{HumanF-MarkrAI,
  title={Gukbap-Gemma3-4B-VL},
  author={MarkrAI},
  year={2025},
  url={https://huggingface.co/HumanF-MarkrAI}
}
```