Dharma Field Extractor

Model Nov 4, 2026

Dharma-AI's reported 2B document model extracts typed JSON fields from page images, with open or schema-directed requests through its documented API.

Visit primary link 2 sources attached

Dharma-AI chart reports Field Extractor at 0.834 versus four tested frontier-model configurations on its document field-extraction evaluation. — Dharma-AI’s own 546-document, 1,638-case evaluation, with low reasoning settings. Scores combine field F1, value similarity and valid JSON; AITrending has not reproduced these results. Source: Dharma-AI

multimodal

What happened lately

Dharma Field Extractor timeline

Launches Nov 4, 2026

Dharma says its 2B field extractor beats frontier models on a document test

Developing 2 sources

Dharma Field Extractor is a specialized document-image model described in Dharma-AI’s November 4 report. The team says it adapts its 2B DharmaOCR-LITE v1.1 base using LoRA and supports open extraction or a caller-defined JSON schema. Its public API documents synchronous image and asynchronous PDF routes, authenticated access and a field-extractor model option. The report’s quality and GPU-cost comparisons are vendor measurements on a bounded benchmark, not independent reproductions or a public per-page sales quote. The reviewed announcement does not establish a public download or license for this particular checkpoint.