Schematron V2 Small

A 3B-parameter model for extracting structured data from HTML and unstructured text into JSON format using JSON schema instructions.

Strengths

Handles extraction from complex schemas and long documents with high fidelity to the defined output structure.

Best for

Reliable schema-driven data extraction tasks where accuracy on intricate page layouts and nested fields matters more than raw speed.

Limitations

Smaller model size and prioritization of extraction quality over speed makes it slower than the newer Schematron V2 Turbo variant for the same workloads.