AI4Bharat

Latest news

1. 16 Oct
   New 2.8T‑token Indic pretraining set and open GPU pipeline announced

   AI4Bharat, IIT Madras and NVIDIA announced Nemotron-pretraining-Indic-v1, a pretraining dataset totaling about 2.8 trillion tokens across 22 Indian languages, and said they will release the GPU‑accelerated data generation pipeline used to create it. The pipeline includes prompts, LLM workflows and quality checks so developers can reproduce or extend the synthetic training data.

   Models · 1 source

   

Articles

No articles about AI4Bharat yet.