G

Gliner Large V2

Developed by urchade
GLiNER is a bidirectional Transformer-based named entity recognition model capable of identifying any entity type, offering greater flexibility than traditional NER models and higher efficiency than large language models.
Downloads 15.73k
Release Time : 3/10/2024

Model Overview

GLiNER is a named entity recognition (NER) model that identifies various entities in text through a bidirectional Transformer encoder. It serves as both an alternative to traditional NER models and a solution to the high resource consumption of large language models.

Model Features

Flexible Entity Recognition
Capable of identifying any user-defined entity types, not limited to predefined entity sets
Efficient Inference
Lower resource consumption and faster inference speed compared to large language models
Business-friendly License
Uses Apache-2.0 license, suitable for commercial applications

Model Capabilities

Text Entity Recognition
Multi-category Entity Extraction
Context-aware Entity Classification

Use Cases

Information Extraction
News Person Identification
Identifying entities such as people, organizations, and locations from news texts
Accurately identifies various named entities in text
Academic Literature Analysis
Extracting specialized terms, method names, and other entities from research papers
Facilitates knowledge graph construction and literature analysis
Business Intelligence
Contract Analysis
Identifying key clauses, dates, and amounts in contracts
Improves contract review efficiency
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase