2B-Parameter OCR Model for Turkish Documents and Benchmark
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
LAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information
MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum
Organization Card
moganai is an initiative working towards an open-source foundation/LLM model for Turkish.
Our goal is to build a high-quality, large-scale pretraining dataset for Turkish, along with the tools we use in the process (data cleaning, quality classification models, etc.), and share them openly with the community. We want to contribute to the growth of the Turkish NLP ecosystem — that's why we release the models and tools we build as open source whenever possible.
MoganBERT Turkish Pre-Trained Encoder Collection
-
moganai/MoganBERT-TR
Fill-Mask • 0.1B • Updated • 169 • 6 -
moganai/MoganBERT-Embed
Sentence Similarity • 0.1B • Updated • 242 • 4 -
moganai/Mogan-ColBERT-TR
Sentence Similarity • 0.1B • Updated • 92 • 4 -
MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum
Paper • 2608.25768 • Published • 5
2B-Parameter OCR Model for Turkish Documents and Benchmark
MoganBERT Turkish Pre-Trained Encoder Collection
-
moganai/MoganBERT-TR
Fill-Mask • 0.1B • Updated • 169 • 6 -
moganai/MoganBERT-Embed
Sentence Similarity • 0.1B • Updated • 242 • 4 -
moganai/Mogan-ColBERT-TR
Sentence Similarity • 0.1B • Updated • 92 • 4 -
MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum
Paper • 2608.25768 • Published • 5
models 12
moganai/MoganOCR-TR
Image-Text-to-Text • 2B • Updated • 175 • 9
moganai/MoganOCR-TR-MLX-4bit
Image-Text-to-Text • 2B • Updated • 25
moganai/MoganOCR-TR-MLX-6bit
Image-Text-to-Text • 2B • Updated • 22
moganai/MoganOCR-TR-MLX-8bit
Image-Text-to-Text • 2B • Updated • 21
moganai/MoganOCR-TR-GGUF
Image-Text-to-Text • 2B • Updated • 262 • 3
moganai/mogan-decision-31B-it
Multiple Choice • Updated • 53 • 1
moganai/lavoir-tr
Text Classification • 0.2B • Updated • 19
moganai/lavoir
Text Classification • 0.4B • Updated • 121
moganai/MoganBERT-Embed
Sentence Similarity • 0.1B • Updated • 242 • 4
moganai/Mogan-ColBERT-TR
Sentence Similarity • 0.1B • Updated • 92 • 4
datasets 8
moganai/TurkishOCR-Bench
Viewer • Updated • 488 • 315
moganai/turkish-synthetic-documents
Viewer • Updated • 200k • 1.24k • 4
moganai/mogan-decision-distill
Viewer • Updated • 123k • 79
moganai/lavoir-dialogues-tr
Viewer • Updated • 57.5k • 232 • 1
moganai/lavoir-dialogues
Viewer • Updated • 32.4k • 191
moganai/mogan-turkish-web
Viewer • Updated • 32.5M • 1.25k • 7
moganai/turkishfineweb2-cleaned
Viewer • Updated • 63.3M • 1.33k • 5
moganai/turkish-resmi-gazete
Viewer • Updated • 20.8k • 127 • 2