AI4Bharat

non-profit

Verified

https://ai4bharat.iitm.ac.in

ai4bharat

AI4Bharat

Activity Feed

AI & ML interests

None defined yet.

Recent Activity

safikhan published a dataset about 18 hours ago

ai4bharat/MILU-1.5

safikhan updated a collection 2 days ago

Indic-LLM-Eval-Suite

safikhan updated a collection 2 days ago

Indic-LLM-Eval-Suite

View all activity

Papers

Seeing Isn't Believing: Uncovering Blind Spots in Evaluator Vision-Language Models

Preferences of a Voice-First Nation: Large-Scale Pairwise Evaluation and Preference Analysis for TTS in Indian Languages

View all Papers

ai4bharat 's collections 18

Bodhan AI Models

Collection of multilingual, multimodal models released in collaboration with Bodhan AI

bodhan-ai/indic-transcribe-core

Automatic Speech Recognition • 1B • Updated 29 days ago • 3.17k • 44
bodhan-ai/indic-transcribe-flex

Automatic Speech Recognition • 1B • Updated 23 days ago • 4.76k • 34
bodhan-ai/indic-ocr

Image-to-Text • Updated Sep 7 • 1.32k • 58
bodhan-ai/indic-translate

Translation • 8B • Updated Sep 7 • 2.42k • 18

Project Astitva

MT, ASR & TTS models for Bhili (भीली), specifically the Dehvali Bhili dialect, an Indo-Aryan language spoken by the Bhil community in western India.

ai4bharat/bhili-translate-bhb-mar

Translation • Updated Sep 1 • 26
ai4bharat/bhili-translate-mar-bhb

Translation • Updated Sep 1 • 55
ai4bharat/bhili-asr-conformer-600m

Automatic Speech Recognition • Updated Sep 1 • 41 • 1
ai4bharat/bhili-asr-conformer-600m-onnx

Automatic Speech Recognition • Updated Sep 1 • 37

Bhojpuri and Hindi Rural Women ASR

This dataset includes ASR data from rural women speaking Hindi and Bhojpuri, supporting inclusive voice recognition.

ai4bharat/Rural_Women_Bhojpuri

Viewer • Updated Jun 15, 2025 • 78.8k • 417 • 6
ai4bharat/Rural_Women_ASR_v2

Viewer • Updated Nov 6, 2025 • 64.4k • 736 • 8

RomanSetu

Romansetu is a collection of models address the challenge of extending Large Language Models (LLMs) to non-English languages using non-Latin scripts

ai4bharat/romansetu-cpt-roman-100m

7B • Updated Mar 7, 2025 • 82
ai4bharat/romansetu-cpt-roman-200m

7B • Updated Mar 7, 2025 • 71
ai4bharat/romansetu-cpt-native-300m

7B • Updated Mar 7, 2025 • 71
ai4bharat/romansetu-cpt-native-400m

7B • Updated Mar 7, 2025 • 74

BhasaAnuvaad

A Speech Translation Dataset for 13 Indian Languages

ai4bharat/WordProject

Viewer • Updated Nov 15, 2024 • 714k • 177 • 5
ai4bharat/Mann-ki-Baat

Viewer • Updated Dec 23, 2024 • 242k • 152 • 6
ai4bharat/NPTEL

Viewer • Updated Nov 15, 2024 • 4.23M • 988 • 11
ai4bharat/UGCE-Resources

Viewer • Updated Feb 4, 2025 • 163k • 94 • 1

CIA Suite

Hercule series of Evaluation models

ai4bharat/hercule-bn

Text Generation • 8B • Updated Oct 18, 2024 • 93 • 1
ai4bharat/hercule-de

Text Generation • 8B • Updated Oct 18, 2024 • 83
ai4bharat/hercule-fr

Text Generation • 8B • Updated Oct 18, 2024 • 84 •
ai4bharat/hercule-hi

Text Generation • 8B • Updated Oct 18, 2024 • 97

IndicLLMSuite

Largest Collections of Pretraining and Instruction Finetuning datasets for 22 Indic languages.

ai4bharat/sangraha

Viewer • Updated Mar 5, 2025 • 268M • 37.3k • 85
ai4bharat/indic-align

Viewer • Updated Jul 25, 2024 • 97.4M • 1.66k • 21
ai4bharat/wiki-translate

Viewer • Updated Jun 24, 2024 • 5.7M • 213 • 8
ai4bharat/wiki-transliterate

Viewer • Updated Jun 24, 2024 • 5.7M • 658 • 4

IndicTrans2

Models(En-Indic, Indic-En, Indic-Indic) in 2 variants (base and dist) and Benchmarks (IN22-Gen and IN22-Conv) released as a part of IndicTrans2.

ai4bharat/indictrans2-en-indic-1B

Translation • 1B • Updated 8 days ago • 68k • 81
ai4bharat/indictrans2-en-indic-dist-200M

Translation • 0.3B • Updated 8 days ago • 239k • 65
ai4bharat/indictrans2-indic-en-1B

Translation • 1B • Updated May 2, 2025 • 9.08k • 56
ai4bharat/indictrans2-indic-en-dist-200M

Translation • 0.2B • Updated 8 days ago • 28.1k • 46

IndicBERT v2

IndicBERT v2 is a multilingual BERT model pretrained on IndicCorp v2, an Indic monolingual corpus of 20.9 billion tokens, covering 24 consitutionally

ai4bharat/IndicBERTv2-MLM-only

Fill-Mask • Updated Sep 5, 2024 • 30.6k • • 10
ai4bharat/IndicBERTv2-MLM-Sam-TLM

Fill-Mask • Updated Sep 5, 2024 • 1.65k • 2
ai4bharat/IndicBERTv2-SS

Fill-Mask • Updated Sep 5, 2024 • 470 • 3
ai4bharat/IndicBERTv2-MLM-Back-TLM

Fill-Mask • Updated Sep 5, 2024 • 292 • 1

Indic-LLM-Eval-Suite

An open, community-driven evolving collection of benchmarks for comprehensive and reliable evaluation of LLMs across Indian Languages

ai4bharat/MILU-1.5

Viewer • Updated 25 days ago • 84.7k • 34
bharatgenai/BhashaBench-Multi

Viewer • Updated Jul 7 • 1.63M • 717 • 6
sarvamai/indivibe

Viewer • Updated May 11 • 4.84k • 62 • 2
ai4bharat/IN22-Gen

Viewer • Updated Mar 19, 2025 • 1.02k • 1.27k • 14

IndicBERT-v3

A collection of state-of-the-art multilingual base encoder language models (270M, 1B, 4B) for Indic languages.

ai4bharat/IndicBERT-v3-1B

Token Classification • 1.0B • Updated Jun 12 • 149 • 9
ai4bharat/IndicBERT-v3-4B

Token Classification • 4B • Updated Jun 12 • 123 • 9
ai4bharat/IndicBERT-v3-270M

Token Classification • 0.3B • Updated Jun 12 • 255 • 4

IndicRagSuite

A comprehensive dataset collection for Indic language information retrieval.

ai4bharat/IndicMSMARCO

Viewer • Updated Jun 3, 2025 • 13k • 212 • 3
ai4bharat/MSMARCO-XI

Viewer • Updated Jun 3, 2025 • 11.5M • 501 • 28

Indic Parler-TTS

Collection of Parler-TTS models adapted to Indian languages.

ai4bharat/indic-parler-tts

Text-to-Speech • 0.9B • Updated 10 days ago • 214k • 421
ai4bharat/indic-parler-tts-pretrained

Text-to-Speech • 0.9B • Updated Dec 9, 2024 • 1.44k • 20
Running on Zero

Agents

Featured

198

Indic Parler-TTS

👀

198

A demo of Indic Parler-TTS

ELAICHI

ELAICHI: Enhancing Low-resource TTS by Addressing Infrequent and Low-frequency Character Bigrams.

ai4bharat/vits-hi-itts-base

Updated Oct 24, 2024 • 1
ai4bharat/vits-hi-proximal-chha

Updated Oct 24, 2024
ai4bharat/vits-hi-synthetic

Updated Oct 24, 2024
ai4bharat/vits-hi-asr-enhanced

Updated Oct 24, 2024

IndicConformer

A collection of ASR models for 22 scheduled languages of India

ai4bharat/indic-conformer-600m-multilingual

Automatic Speech Recognition • Updated 8 days ago • 434k • 229
ai4bharat/indicconformer_stt_as_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 74 • 4
ai4bharat/indicconformer_stt_bn_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 331 • 2
ai4bharat/indicconformer_stt_brx_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 4

Airavata Evaluation Suite

A collection of benchmarks used for evaluation of Airavata, an Hindi instruction-tuned model on top of Sarvam's OpenHathi base model.

ai4bharat/IndicSentiment

Updated May 6, 2024 • 1.22k • 11
ai4bharat/IndicSentiment-Translated

Viewer • Updated Oct 5, 2023 • 1.16k • 74 • 2
ai4bharat/IndicCOPA

Updated Dec 15, 2022 • 525 • 3
ai4bharat/IndicCOPA-Translated

Viewer • Updated Jan 3, 2024 • 449 • 22 • 1

IndicXTREME

IndicXTREME is a human-supervised benchmark of 9 diverse NLU tasks across 20 languages, featuring 105 evaluation sets in total.

ai4bharat/IndicSentiment

Updated May 6, 2024 • 1.22k • 11
ai4bharat/IndicCOPA

Updated Dec 15, 2022 • 525 • 3
ai4bharat/IndicXParaphrase

Viewer • Updated Oct 14, 2022 • 20k • 312 • 5
Divyanshu/indicxnli

Viewer • Updated Oct 6, 2022 • 4.4M • 1.01k • 8

IndicNLG

IndicNLG Benchmark is a dataset collection designed for benchmarking Natural Language Generation (NLG) across 11 Indic languages.

ai4bharat/IndicParaphrase

Viewer • Updated Oct 13, 2022 • 5.56M • 1.22k • 6
ai4bharat/IndicWikiBio

Viewer • Updated Oct 13, 2022 • 57.4k • 846 • 2
ai4bharat/IndicQuestionGeneration

Viewer • Updated Oct 13, 2022 • 1.08M • 343 • 5
ai4bharat/IndicSentenceSummarization

Viewer • Updated Oct 13, 2022 • 1.11M • 241 • 6

Bodhan AI Models

Collection of multilingual, multimodal models released in collaboration with Bodhan AI

bodhan-ai/indic-transcribe-core

Automatic Speech Recognition • 1B • Updated 29 days ago • 3.17k • 44
bodhan-ai/indic-transcribe-flex

Automatic Speech Recognition • 1B • Updated 23 days ago • 4.76k • 34
bodhan-ai/indic-ocr

Image-to-Text • Updated Sep 7 • 1.32k • 58
bodhan-ai/indic-translate

Translation • 8B • Updated Sep 7 • 2.42k • 18

Indic-LLM-Eval-Suite

An open, community-driven evolving collection of benchmarks for comprehensive and reliable evaluation of LLMs across Indian Languages

ai4bharat/MILU-1.5

Viewer • Updated 25 days ago • 84.7k • 34
bharatgenai/BhashaBench-Multi

Viewer • Updated Jul 7 • 1.63M • 717 • 6
sarvamai/indivibe

Viewer • Updated May 11 • 4.84k • 62 • 2
ai4bharat/IN22-Gen

Viewer • Updated Mar 19, 2025 • 1.02k • 1.27k • 14

Project Astitva

MT, ASR & TTS models for Bhili (भीली), specifically the Dehvali Bhili dialect, an Indo-Aryan language spoken by the Bhil community in western India.

ai4bharat/bhili-translate-bhb-mar

Translation • Updated Sep 1 • 26
ai4bharat/bhili-translate-mar-bhb

Translation • Updated Sep 1 • 55
ai4bharat/bhili-asr-conformer-600m

Automatic Speech Recognition • Updated Sep 1 • 41 • 1
ai4bharat/bhili-asr-conformer-600m-onnx

Automatic Speech Recognition • Updated Sep 1 • 37

IndicBERT-v3

A collection of state-of-the-art multilingual base encoder language models (270M, 1B, 4B) for Indic languages.

ai4bharat/IndicBERT-v3-1B

Token Classification • 1.0B • Updated Jun 12 • 149 • 9
ai4bharat/IndicBERT-v3-4B

Token Classification • 4B • Updated Jun 12 • 123 • 9
ai4bharat/IndicBERT-v3-270M

Token Classification • 0.3B • Updated Jun 12 • 255 • 4

Bhojpuri and Hindi Rural Women ASR

This dataset includes ASR data from rural women speaking Hindi and Bhojpuri, supporting inclusive voice recognition.

ai4bharat/Rural_Women_Bhojpuri

Viewer • Updated Jun 15, 2025 • 78.8k • 417 • 6
ai4bharat/Rural_Women_ASR_v2

Viewer • Updated Nov 6, 2025 • 64.4k • 736 • 8

IndicRagSuite

A comprehensive dataset collection for Indic language information retrieval.

ai4bharat/IndicMSMARCO

Viewer • Updated Jun 3, 2025 • 13k • 212 • 3
ai4bharat/MSMARCO-XI

Viewer • Updated Jun 3, 2025 • 11.5M • 501 • 28

RomanSetu

Romansetu is a collection of models address the challenge of extending Large Language Models (LLMs) to non-English languages using non-Latin scripts

ai4bharat/romansetu-cpt-roman-100m

7B • Updated Mar 7, 2025 • 82
ai4bharat/romansetu-cpt-roman-200m

7B • Updated Mar 7, 2025 • 71
ai4bharat/romansetu-cpt-native-300m

7B • Updated Mar 7, 2025 • 71
ai4bharat/romansetu-cpt-native-400m

7B • Updated Mar 7, 2025 • 74

Indic Parler-TTS

Collection of Parler-TTS models adapted to Indian languages.

ai4bharat/indic-parler-tts

Text-to-Speech • 0.9B • Updated 10 days ago • 214k • 421
ai4bharat/indic-parler-tts-pretrained

Text-to-Speech • 0.9B • Updated Dec 9, 2024 • 1.44k • 20
Running on Zero

Agents

Featured

198

Indic Parler-TTS

👀

198

A demo of Indic Parler-TTS

BhasaAnuvaad

A Speech Translation Dataset for 13 Indian Languages

ai4bharat/WordProject

Viewer • Updated Nov 15, 2024 • 714k • 177 • 5
ai4bharat/Mann-ki-Baat

Viewer • Updated Dec 23, 2024 • 242k • 152 • 6
ai4bharat/NPTEL

Viewer • Updated Nov 15, 2024 • 4.23M • 988 • 11
ai4bharat/UGCE-Resources

Viewer • Updated Feb 4, 2025 • 163k • 94 • 1

ELAICHI

ELAICHI: Enhancing Low-resource TTS by Addressing Infrequent and Low-frequency Character Bigrams.

ai4bharat/vits-hi-itts-base

Updated Oct 24, 2024 • 1
ai4bharat/vits-hi-proximal-chha

Updated Oct 24, 2024
ai4bharat/vits-hi-synthetic

Updated Oct 24, 2024
ai4bharat/vits-hi-asr-enhanced

Updated Oct 24, 2024

CIA Suite

Hercule series of Evaluation models

ai4bharat/hercule-bn

Text Generation • 8B • Updated Oct 18, 2024 • 93 • 1
ai4bharat/hercule-de

Text Generation • 8B • Updated Oct 18, 2024 • 83
ai4bharat/hercule-fr

Text Generation • 8B • Updated Oct 18, 2024 • 84 •
ai4bharat/hercule-hi

Text Generation • 8B • Updated Oct 18, 2024 • 97

IndicConformer

A collection of ASR models for 22 scheduled languages of India

ai4bharat/indic-conformer-600m-multilingual

Automatic Speech Recognition • Updated 8 days ago • 434k • 229
ai4bharat/indicconformer_stt_as_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 74 • 4
ai4bharat/indicconformer_stt_bn_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 331 • 2
ai4bharat/indicconformer_stt_brx_hybrid_ctc_rnnt_large

Automatic Speech Recognition • Updated Nov 4, 2024 • 4

IndicLLMSuite

Largest Collections of Pretraining and Instruction Finetuning datasets for 22 Indic languages.

ai4bharat/sangraha

Viewer • Updated Mar 5, 2025 • 268M • 37.3k • 85
ai4bharat/indic-align

Viewer • Updated Jul 25, 2024 • 97.4M • 1.66k • 21
ai4bharat/wiki-translate

Viewer • Updated Jun 24, 2024 • 5.7M • 213 • 8
ai4bharat/wiki-transliterate

Viewer • Updated Jun 24, 2024 • 5.7M • 658 • 4

Airavata Evaluation Suite

A collection of benchmarks used for evaluation of Airavata, an Hindi instruction-tuned model on top of Sarvam's OpenHathi base model.

ai4bharat/IndicSentiment

Updated May 6, 2024 • 1.22k • 11
ai4bharat/IndicSentiment-Translated

Viewer • Updated Oct 5, 2023 • 1.16k • 74 • 2
ai4bharat/IndicCOPA

Updated Dec 15, 2022 • 525 • 3
ai4bharat/IndicCOPA-Translated

Viewer • Updated Jan 3, 2024 • 449 • 22 • 1

IndicTrans2

Models(En-Indic, Indic-En, Indic-Indic) in 2 variants (base and dist) and Benchmarks (IN22-Gen and IN22-Conv) released as a part of IndicTrans2.

ai4bharat/indictrans2-en-indic-1B

Translation • 1B • Updated 8 days ago • 68k • 81
ai4bharat/indictrans2-en-indic-dist-200M

Translation • 0.3B • Updated 8 days ago • 239k • 65
ai4bharat/indictrans2-indic-en-1B

Translation • 1B • Updated May 2, 2025 • 9.08k • 56
ai4bharat/indictrans2-indic-en-dist-200M

Translation • 0.2B • Updated 8 days ago • 28.1k • 46

IndicXTREME

IndicXTREME is a human-supervised benchmark of 9 diverse NLU tasks across 20 languages, featuring 105 evaluation sets in total.

ai4bharat/IndicSentiment

Updated May 6, 2024 • 1.22k • 11
ai4bharat/IndicCOPA

Updated Dec 15, 2022 • 525 • 3
ai4bharat/IndicXParaphrase

Viewer • Updated Oct 14, 2022 • 20k • 312 • 5
Divyanshu/indicxnli

Viewer • Updated Oct 6, 2022 • 4.4M • 1.01k • 8

IndicBERT v2

IndicBERT v2 is a multilingual BERT model pretrained on IndicCorp v2, an Indic monolingual corpus of 20.9 billion tokens, covering 24 consitutionally

ai4bharat/IndicBERTv2-MLM-only

Fill-Mask • Updated Sep 5, 2024 • 30.6k • • 10
ai4bharat/IndicBERTv2-MLM-Sam-TLM

Fill-Mask • Updated Sep 5, 2024 • 1.65k • 2
ai4bharat/IndicBERTv2-SS

Fill-Mask • Updated Sep 5, 2024 • 470 • 3
ai4bharat/IndicBERTv2-MLM-Back-TLM

Fill-Mask • Updated Sep 5, 2024 • 292 • 1

IndicNLG

IndicNLG Benchmark is a dataset collection designed for benchmarking Natural Language Generation (NLG) across 11 Indic languages.

ai4bharat/IndicParaphrase

Viewer • Updated Oct 13, 2022 • 5.56M • 1.22k • 6
ai4bharat/IndicWikiBio

Viewer • Updated Oct 13, 2022 • 57.4k • 846 • 2
ai4bharat/IndicQuestionGeneration

Viewer • Updated Oct 13, 2022 • 1.08M • 343 • 5
ai4bharat/IndicSentenceSummarization

Viewer • Updated Oct 13, 2022 • 1.11M • 241 • 6

AI & ML interests

Recent Activity

Papers

Team members 36

ai4bharat 's collections 18

Indic Parler-TTS

Indic Parler-TTS