Skip to content
assaynotes
EN日本語中文

Hugging Face releases Dharma field extraction model

Developing1 reports1 sourcesupdated 4 h ago

What happened

AI digest

On 5 November 2026, Dharma-AI published Dharma Field Extractor on the Hugging Face blog. According to the report, it is a 2-billion-parameter model that extracts specified fields from any type of document and returns them as JSON, with two modes: open extraction without a schema, and targeted extraction following a given schema. The report gives no further details such as training data, evaluation results, licence or launch date; the story remains at the release stage, with further developments still to be seen.

Written by AI from the reports · updated 2 h ago

LatestNov 5 01:50
Dharma-AI releases the 2-billion-parameter Dharma Field Extractor, supporting open and schema-based extraction with JSON output.

Timeline

Follow the reports to see the event from each side.

Nov 5
  1. Hugging Face Blog
    Dharma Field Extractor: Any Document In, Structured Fields in JSON Out

    A 2-billion-parameter model that extracts exactly the fields you ask for, from any document type, outperforming every frontier LLM we tested, at a fraction of the cost.

Heat of this event

Not enough continuous observations yet to draw a trend.