LLM Made by the Indians, for the Indians

Can't read or see images? View this email in a browser

TAUSIF ALAM & AMIT RAJA NAIK Wednesday, Jan 24, 2024 | Was this email forwarded to you? Sign up here ___________________________________________________________ The US has GPT, France is equipped with LLM models like LLaMA and Mistral, and China boasts of Ernie in its fourth phase; the only eminent player missing in the LLM race was India – until BharatGPT, the country’s own Indic LLM. However, the bigger story is that it will be the Meta of Indic LLMs. The makers of BharatGPT aim to build a foundational model like LLaMA, which will be the biggest contributor to the open source community. The initial model will be open source, inviting developers to contribute to its enhancement. The belief is that the more widespread its usage, the more improvements it will undergo. The plan is to release additional versions of the model eventually. The makers have already started open-sourcing a lot of its work on Decile (similar to GitHub and Hugging Face). Remarkably, the model is based on Indian languages. The makers stress that GPT-4 and any other English language-based foundational models won’t be able to make a GPT-4-like model for languages such as Kannada, Tamil, or Hindi. It has to come from India alone.  The BharatGPT team is developing video and speech models, which are set to be released shortly after the launch of the initial foundational text-based model. Read the full story here. OpenAI Catches ’em young  OpenAI is actively fishing for young minds, trying to catch those who can help them in their quest for continuous innovation. The company runs a six-month knowledge program called OpenAI Residency that aims to provide critical skills to researchers and engineers in other fields and help them transition into the AI & ML space. It also has an online community named ‘Young people in artificial intelligence’, where users can discuss AI and OpenAI-related topics.  Read the full story here.  Infosys Debuts in Australian Open 2024    Infosys is intensifying its AI presence at the 2024 Australian Open, utilising generative AI technologies to boost fan engagement, optimise player performance, and facilitate digital content creation. Leveraging Infosys Topaz, the company offers services like the AO 2024 Bracket Challenge vs AI, Infosys Match Centre with contextual insights, AI Match Bytes for visual match story cards, and a Fan Zone featuring generative AI graffiti portraits.  The partnership aims to enhance the Australian Grand Slam experience while promoting digital skilling platforms for future leaders in Australia. Last year, Infosys also signed Rafael Nadal as its brand ambassador. Read the full story here.  Jensen Cameos at Beijing Party NVIDIA chief Jensen Huang made a surprise appearance at the 2024 Beijing New Year Party, marking his first visit to China in four years. Giving his usual black leather jacket a break, Huang donned a traditional red vest and even performed for NVIDIA employees.  The visit follows Huang's low-key visits to offices in Beijing, Shanghai, and Shenzhen. Amidst US-China geopolitical tensions, including a ban on AI chip exports, Huang's strategic and fun-filled visit signals a unique engagement with Chinese facilities. Read the full story here.   Register for our Upcoming Workshop - Building Generative AI Applications with Amazon Bedrock DOWNLOAD OUR MOBILE APP Stay Connected info@analyticsindiamag.com © 2023 Analytics India Magazine

This email was sent by support@analyticsindiamag.com to alexvarboffin.abbb@blogger.com Not interested? Unsubscribe | Manage Preference | Update profile Analytics India Magazine | 280, 2nd floor, 5th Main, 15 A cross, Sector 6, HSR layout Bengaluru, Karnataka 560102