gliner2-base-v1

Name: gliner2-base-v1
Author: fastino

by fastino

Unified NER, classification, and structured extraction in a single 205M CPU-efficient model

379Kdl/month

205Mparams

HuggingFace Run it on your own data, free

Identifiers

Model ID

fastino/gliner2-base-v1

Feature URI

mixpeek://document_extractor@v1/fastino_gliner2_base_v1

Overview

GLiNER 2 unifies named entity recognition, text classification, and hierarchical structured data extraction into a single 205M-parameter model built on a pretrained transformer encoder. Unlike pipeline approaches that chain separate models or LLM-based extraction that requires GPU infrastructure, GLiNER 2 runs efficiently on CPU with an intuitive schema-based interface that accepts natural language type descriptions.

On Mixpeek, GLiNER 2 powers lightweight entity extraction pipelines that run alongside heavier models without competing for GPU resources. Its zero-shot generalization across domains (matching GPT-4o on CrossNER benchmarks) makes it ideal for extracting custom entities from transcripts, OCR output, and document text without fine-tuning.

Architecture

Pretrained transformer encoder with multi-task composition heads for NER, classification, and structured extraction. 205M parameters. Schema-driven interface supporting natural language entity type descriptions, nested and overlapping spans, and configurable single or multi-label classification.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so gliner2-base-v1 runs
// on your side and the output is upserted through POST
// /v1/namespaces/{namespace_id}/documents/upsert. On Enterprise the other
// path is to upload the weights instead: POST /v1/namespaces/{id}/models
// accepts the huggingface format and a custom plugin loads them.
const res = await fetch(
  "https://api.mixpeek.com/v1/namespaces/ns_your_namespace/documents/upsert",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      collection_id: "col_your_collection",
      documents: [
        {
          document_id: "asset-00412",
          // The model produces text, so it lands in payload. Give the
          // collection a text vector index and embed that text to make it
          // searchable rather than only filterable.
          payload: { extracted_text: modelOutput, source_key: "archive/2026/asset-00412" },
          vectors: { "text-embedding": embeddingOfModelOutput },
        },
      ],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// universal_extractor@v1 runs google/gemini-embedding-2
// (3072-d) over a bucket, with no inference of your own.