Ba2han/Turkish-OCR-test
The Ba2han/Turkish-OCR-test is a 2.3 billion parameter Qwen3.5-based language model developed by Ba2han, fine-tuned from unsloth/Qwen3.5-2B. This model is specifically optimized for tasks related to Turkish OCR, leveraging efficient training with Unsloth and Huggingface's TRL library. It offers a context length of 32768 tokens, making it suitable for processing extensive Turkish text data.
Loading preview...
Model Overview
The Ba2han/Turkish-OCR-test is a 2.3 billion parameter language model developed by Ba2han. It is a Qwen3.5-based model, specifically fine-tuned from the unsloth/Qwen3.5-2B base model. This model was trained with a focus on efficiency, utilizing the Unsloth library, which enabled a 2x faster training process, in conjunction with Huggingface's TRL library.
Key Characteristics
- Architecture: Based on the Qwen3.5 family.
- Parameter Count: 2.3 billion parameters.
- Training Efficiency: Leverages Unsloth for accelerated fine-tuning.
- Context Length: Supports a substantial context window of 32768 tokens.
Primary Use Case
This model is specifically designed and fine-tuned for tasks related to Turkish Optical Character Recognition (OCR). Its optimization and training methodology suggest strong performance in processing and understanding Turkish text, particularly in scenarios where OCR output needs further linguistic processing or analysis.