🏗️ Building on HF

Göktuğ Düşünen PRO

GoktugD

11 3

https://werea.co

AI & ML interests

Head of AI @ Werea · AI/ML Engineer AI/ML Engineer building end-to-end intelligent systems across LLMs, NLP, Computer Vision, RAG, Retrieval, Agents, and Applied AI. From data and model development to fine-tuning, evaluation, optimization, deployment, and production AI systems.

Recent Activity

new activity 2 days ago

Werea-co/Werea-TR-TextRestore:Demo for this model on Spaces

new activity 2 days ago

Werea-co/Werea-TR-TextRestore:Demo for this model on Spaces

posted an update 2 days ago

🇹🇷 **A 1B Turkish OCR model vs. Baidu OCR.** We ran both models on the **same Turkish enterprise documents**. The result: **Werea-DocOCR-1B → 99.9** **Baidu Unlimited-OCR → 47.9** Same documents. Same evaluation. And Werea-DocOCR is only **1B parameters**. It was built specifically for difficult Turkish enterprise documents: 📄 invoices 📑 contracts 🏦 bank receipts 💼 payroll 🚗 vehicle documents 📋 SGK-style tables 📱 scanned & phone-captured documents But benchmarks aren't enough. **Give me a Turkish document that you think will break it.** We'll test the hardest ones and publish the failures. 🤗 Model: https://huggingface.co/Werea-co/Werea-DocOCR-1B 📚 Dataset: https://huggingface.co/datasets/Werea-co/werea-tr-doc-ocr-enterprise-v2 🇹🇷 Built in Türkiye. Open on Hugging Face. #OCR #DocumentAI #HuggingFace #TurkishAI

View all activity

Organizations

Posts 7

Post

3086

🇹🇷 **A 1B Turkish OCR model vs. Baidu OCR.**

We ran both models on the **same Turkish enterprise documents**.

The result:

**Werea-DocOCR-1B → 99.9**
**Baidu Unlimited-OCR → 47.9**

Same documents.
Same evaluation.

And Werea-DocOCR is only **1B parameters**.

It was built specifically for difficult Turkish enterprise documents:

📄 invoices
📑 contracts
🏦 bank receipts
💼 payroll
🚗 vehicle documents
📋 SGK-style tables
📱 scanned & phone-captured documents

But benchmarks aren't enough.

**Give me a Turkish document that you think will break it.**

We'll test the hardest ones and publish the failures.

🤗 Model:
Werea-co/Werea-DocOCR-1B

📚 Dataset:
Werea-co/werea-tr-doc-ocr-enterprise-v2

🇹🇷 Built in Türkiye. Open on Hugging Face.

#OCR #DocumentAI #HuggingFace #TurkishAI

Post

🇹🇷 **You tried to break our Turkish NER model.**

And that's exactly what we wanted.

After sharing Werea-TR-NER, we asked the Hugging Face community to challenge our ~110M parameter model with difficult Turkish sentences.

Some worked.

Some exposed weaknesses.

And that's more valuable than pretending a benchmark score tells the whole story.

Our published WikiANN-tr result:

**91.7% Entity F1**

👤 PERSON → 94.2%
📍 LOCATION → 91.4%
🏢 ORGANIZATION → 89.2%

But now we want to go further.

🔥 **ROUND 2**

Send me a Turkish sentence designed specifically to break the model.

Ambiguous names.
Companies that sound like people.
Locations hidden inside organization names.
Turkish suffixes.
Slang.
Anything nasty.

**Try to make it fail.**

I'll collect the hardest examples and use them to build a public adversarial Turkish NER evaluation set.

🤗 Model:
Werea-co/Werea-TR-NER

🇹🇷 Werea:

Werea-co

Follow me if you want to see whether the community can break it — and what we build from the failures.

#TurkishNLP #HuggingFace #NER #OpenSourceAI