AI4Bharat · India · Released 2024-05

IndicTrans2 / AI4Bharat Suite

Open translation and understanding models from IIT Madras covering all 22 scheduled Indian languages.

Official site GitHub

Access

open

Context

4,096 tokens

Modalities

text

Languages

22 Indian languages

Benchmark scores

MMLU

N/A

Massive Multitask Language Understanding

GPQA

N/A

Graduate-level science QA (Diamond)

HumanEval

N/A

Python code completion

SWE-bench

N/A

Real GitHub issue resolution (Verified)

LiveCodeBench

N/A

Contamination-free coding

AIME

N/A

American Invitational Mathematics Exam

MMMU

N/A

Multimodal college-level reasoning

MATH

N/A

Competition-level math word problems

Scores as reported by the vendor or leading public leaderboards. "N/A" means the score has not been publicly disclosed for this metric.

Explain like I'm 5

Open research models for translating and understanding Indian languages.

Key features

Fully open
22 languages
Translation + STT + TTS

Strengths

Open license
Broad Indic coverage

Limitations

Not a chat LLM

Best for

TranslationSpeech research

Related models

OpenAI

GPT-5

OpenAI's flagship reasoning + multimodal model with a unified router that switches between fast and deep-thinking modes.

Google DeepMind

Gemini 2.5 Pro

Google's most capable model with native 1M-token context and full multimodal I/O across text, images, audio and video.

Meta AI

Llama 4 Maverick

Meta's flagship MoE model — 400B total / 17B active — with native multimodal input and 1M-token context.

Alibaba Qwen

Qwen3-Max

Alibaba's flagship trillion-parameter Qwen3 tier — strong multilingual, coding and agentic performance.