DharmaOCR: Open-Source Specialized SLM (3B) + Cost–Performance Benchmark against LLMs and other open-sourced models [R]
Hey everyone, we just open-sourced DharmaOCR on Hugging Face. Models and datasets are all public, free to use and experiment with. We also published the paper documenting all the experimentation behind it, for those who want to dig into the methodology…