Back to all models

Unable to determine this model’s pipeline type. Check the docs .

Monthly model downloads

dbmdz/distilbert-base-turkish-cased dbmdz/distilbert-base-turkish-cased
last 30 days



Contributed by

Bayerische Staatsbibliothek non-profit
1 team member Β· 24 models

How to use this model directly from the πŸ€—/transformers library:

Copy to clipboard
from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("dbmdz/distilbert-base-turkish-cased") model = AutoModel.from_pretrained("dbmdz/distilbert-base-turkish-cased")

πŸ€— + πŸ“š dbmdz Distilled Turkish BERT model

In this repository the MDZ Digital Library team (dbmdz) at the Bavarian State Library open sources a (cased) distilled model for Turkish πŸŽ‰

πŸ‡ΉπŸ‡· DistilBERTurk

DistilBERTurk is a community-driven cased distilled BERT model for Turkish.

DistilBERTurk was trained on 7GB of the original training data that was used for training BERTurk, using the cased version of BERTurk as teacher model.

DistilBERTurk was trained with the official Hugging Face implementation from here for 5 days on 4 RTX 2080 TI.

More details about distillation can be found in the "DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter" paper by Sanh et al. (2019).

Model weights

Currently only PyTorch-Transformers compatible weights are available. If you need access to TensorFlow checkpoints, please raise an issue in the BERTurk repository!

Model Downloads
dbmdz/distilbert-base-turkish-cased config.json β€’ pytorch_model.bin β€’ vocab.txt


With Transformers >= 2.3 our DistilBERTurk model can be loaded like:

from transformers import AutoModel, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("dbmdz/distilbert-base-turkish-cased")
model = AutoModel.from_pretrained("dbmdz/distilbert-base-turkish-cased")


For results on PoS tagging or NER tasks, please refer to this repository.

For PoS tagging, DistilBERTurk outperforms the 24-layer XLM-RoBERTa model.

The overall performance difference between DistilBERTurk and the original (teacher) BERTurk model is ~1.18%.

Huggingface model hub

All models are available on the Huggingface model hub.

Contact (Bugs, Feedback, Contribution and more)

For questions about our BERT models just open an issue here πŸ€—


Thanks to Kemal Oflazer for providing us additional large corpora for Turkish. Many thanks to Reyyan Yeniterzi for providing us the Turkish NER dataset for evaluation.

Research supported with Cloud TPUs from Google's TensorFlow Research Cloud (TFRC). Thanks for providing access to the TFRC ❀️

Thanks to the generous support from the Hugging Face team, it is possible to download both cased and uncased models from their S3 storage πŸ€—