IndicGLUE Benchmark

IndicGLUE Benchmark

MH Specific

Marathi Subset — Natural language understanding benchmark for 11 Indian languages including Marathi, covering tasks like news categorisation, headline prediction, and paraphrase detection

Run comprehensive NLU benchmarks on Marathi models using IndicGLUE to identify areas needing improvement.
HomepageHuggingFace

Quick Start

from datasets import load_dataset
ds = load_dataset('ai4bharat/indic_glue', 'wnli.mr')
print(f'Tasks available for Marathi (mr)')
print(f'WNLI samples: {len(ds["train"])}')
Modality
Text (Marathi)
Size
Multi-task, Marathi subset
License
Format
CSV/JSON
Language
mr
Update Frequency
static
Organization
AI4Bharat, IIT Madras
Category
benchmarks-fairness-dialects-tools

Schema

FieldTypeDescription
taskstringNLU task name
textstringInput text for the task
labelstringTask-specific label

Build With This

Create an automated Marathi model evaluation suite that benchmarks new models on all IndicGLUE tasks
Develop a model selection guide recommending the best Marathi NLP model for each specific task type
Build a Marathi NLU leaderboard website tracking model performance over time on IndicGLUE benchmarks

AI Use Cases

NLU model evaluationMarathi language model benchmarking
Last verified: 2026-03-07