---
language:
- "th"
tags:
- "thai"
- "masked-lm"
- "modernbert"
datasets:
- "wikimedia/wikipedia"
license: "apache-2.0"
pipeline_tag: "fill-mask"
mask_token: "[MASK]"
---

# modernbert-base-thai-wikipedia

## Model Description

This is a ModernBERT model pre-trained on Thai Wikipedia texts. NVIDIA A100-SXM4-40GB×8 took 3 hours 9 minutes for training. You can fine-tune `modernbert-base-thai-wikipedia` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/modernbert-base-thai-wikipedia-upos), [dependency-parsing](https://huggingface.co/KoichiYasuoka/modernbert-base-thai-wikipedia-ud-embeds), and so on.

## How to Use

```py
from transformers import AutoTokenizer,AutoModelForMaskedLM
tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/modernbert-base-thai-wikipedia")
model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/modernbert-base-thai-wikipedia",trust_remote_code=True)
```