theaicatalog
Hugging Face ยท Models

Dharma-AI releases a 2-billion-parameter model that pulls fields from documents as JSON

Dharma-AI has released Field Extractor, a model that takes a document image and returns structured JSON. Users can either ask for every field the model finds or supply an exact schema, and fields missing from the document come back as null. The model is built on DharmaOCR-LITE v1.1 with a LoRA adapter trained on 43,384 examples, and in the company's 546-document benchmark it scored 0.834, above Claude Opus 5 at 0.816 and GPT-5.6 Luna at 0.741. The company says it runs on a single L40S at a lower cost per extraction than every API it tested.