Dharma AI
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 2
- Dharma AI releases 2B Field Extractor for structured JSON
Dharma AI released the 2-billion-parameter Dharma Field Extractor, which accepts document images and returns typed fields as structured JSON. It supports open-ended extraction without a schema and schema-directed extraction with named, typed fields, including nested invoice line items, without per-document-type fine-tuning. Dharma AI reports that the model beat four general-purpose frontier configurations on its benchmark and cost 6 to 147 times less, but these comparisons are vendor-reported and depend on the company’s evaluation setup.
Hugging Face · 🔥 0 - VocabX Cuts Inference Cost 77% by Rebuilding Vocabulary for Domain Models
Dharma AI, in a Hugging Face article, previewed VocabX, a method that rebuilds a language model's tokenizer around the cost of each token rather than how often it appears, combining a domain-specific sub-word vocabulary with multi-token macros for repeated spans such as JSON envelopes. The authors say they applied it to a production Portuguese essay grader built on Qwen3-4B, keeping the same base model, data and training recipe, and cut measured inference cost by 77% while shrinking the output layer from about 152,000 to about 65,000 rows. They also report that VocabX beat fast vocabulary transfer, multi-word tokenization and Liquid AI's in-place BPE across 15 domain and model combinations (contracts, medical papers, scientific papers, news and Python code on Qwen3, Llama-3.1 and Gemma-3). The results come from a VocabX paper that Dharma AI says is still under review and may change before publication.
Hugging Face · 🔥 0
Experience and discussion from the community
Share my Dharma AI experienceAsk about Dharma AI
Nobody has shared their experience with Dharma AI yet.