Dharma-AI Releases 2B Document Field Extraction Model Field Extractor
Understand the story
On November 5, 2026, Dharma-AI released the 2-billion-parameter Field Extractor model via the Hugging Face Blog. The model targets document field extraction scenarios: it takes document images as input and outputs structured JSON, supporting two working modes—an open mode and a schema-directed mode. For fields that do not exist in a document, the model returns null instead of fabricating content, reducing hallucinated information in extraction results. This is the first report on the event, which is still developing.
AI overview of reporting · Updated 1 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- Hugging Face BlogDharma Field Extractor: Any Document In, Structured Fields in JSON Out
Dharma-AI released Field Extractor, a 2-billion-parameter model that takes document images as input and outputs structured JSON in two modes—open-ended or schema-directed—returning null for missing fields instead of fabricating values.
Story heat trend
Not enough continuous observations to show a trend yet.