AI4Bharat Sangraha
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
AI4Bharat Sangraha Total Indic Corpus
6,623Token Count (M)
3
AI4Bharat Sangraha Telugu
599Token Count (M)
3
AI4Bharat Sangraha Tamil
684Token Count (M)
3
AI4Bharat Sangraha Punjabi
210Token Count
3
AI4Bharat Sangraha Odia
228Token Count (M)
3
AI4Bharat Sangraha Marathi
529,000,000Token Count
3
AI4Bharat Sangraha Malayalam
518Token Count
3
AI4Bharat Sangraha Kannada
313Token Count (M)
3
AI4Bharat Sangraha Hindi
1,231,000,000Token Count (M)
3
AI4Bharat Sangraha Gujarati
605Token Count (M)
3
AI4Bharat Sangraha Bengali
1,638Token Count (M)
3
AI4Bharat Sangraha Assamese
100Token Count (M)
3