I

Indobert Lite Base P1

Developed by indobenchmark
IndoBERT is a BERT model variant tailored for the Indonesian language, trained using masked language modeling and next sentence prediction objectives. The Lite version is a lightweight model suitable for resource-constrained environments.
Downloads 723
Release Time : 3/2/2022

Model Overview

An Indonesian pretrained language model optimized based on the BERT architecture, focusing on natural language understanding tasks, offering both base and lightweight versions.

Model Features

Indonesian Language Optimization
Specially pretrained and optimized for Indonesian language characteristics
Lightweight Design
The Lite version significantly reduces parameters, making it suitable for resource-limited scenarios
Two-Phase Training
Offers P1 (Case Insensitive) and P2 (Case Sensitive) versions

Model Capabilities

Indonesian Text Understanding
Contextual Feature Extraction
Masked Word Prediction

Use Cases

Natural Language Processing
Text Classification
Classification of Indonesian news/articles
Named Entity Recognition
Entity recognition in Indonesian text
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase