Skip to content

Multi-task Indic Language Understanding Benchmark

Culturally grounded knowledge comprehension across ten Indic languages and English.

3 models

Higher is better.

  1. 1Claude Opus 5.5Anthropic93.1
  2. 2Claude Opus 5Anthropic92.1
  3. 3Claude Sonnet 5.5Anthropic91.6
Higher is better. Bar colour: blue proprietary, orange open weights, green other.

About this benchmark

Metric
Average accuracy
Tasks
Knowledge tasks across 11 languages
Authors
Verma et al. (2024)

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.