Dharma Field Extractor Turns Document Images into Structured JSON
Dharma Field Extractor: Any Document In, Structured Fields in JSON Out
- Who
- Dharma-AI
- What
- releases a 2-billion-parameter model for extracting structured fields from documents
- Quote
So we built a model for it instead.
What happened
Dharma-AI introduces the Dharma Field Extractor, a 2-billion-parameter model that converts document images into structured JSON, either by extracting fields it identifies or by filling a user-specified schema. The authors report a score of 0.834 across 1,638 benchmark cases, above four tested frontier configurations, with self-hosted GPU costs 6 to 147 times lower than the compared APIs.
Why it matters
The benchmark compares field extraction quality and operating costs across a specialized 2-billion-parameter model and four frontier configurations, showing how the authors evaluate specialization on this task.
Read next
See the full storyDharma-AI Releases a 2-Billion-Parameter Document Field Extraction ModelThe story so far
Dharma-AI introduced the Dharma Field Extractor, a 2-billion-parameter model that turns document images into structured JSON by either extracting fields it identifies or filling a schema supplied by the user. The model is now available for those two extraction workflows. The authors reported a score of 0.834 across 1,638 benchmark cases, higher than four frontier configurations they tested. They also reported that self-hosting the model on GPUs cost 6 to 147 times less than the compared APIs.
Keep up with AI, one email a day
Model releases, price changes and compute developments, with original sources. Free, confirmed by email, and easy to unsubscribe. Read a past issue · 2026-11-04