<\/p>
BharatGen on February 16 launched Param 2, a 17-billion-parameter artificial intelligence model designed to serve India\u2019s linguistic and sectoral needs, positioning it as a domestic alternative to global systems from companies such as OpenAI and Google.<\/p>
The model, unveiled at the India AI Impact Summit 2026, supports 22 Indian languages.<\/p>
\u201cParam 2 launches today, that\u2019s our 17-billion-parameter mixture-of-experts model, 22 Indian languages, support for code, maths and reasoning,\u201d BharatGen Chief Executive Rishi Bal told Business Today <\/em>on the sidelines of the summit. \u201cIt\u2019s a pre-trained checkpoint\u2026 developers can take it and fine-tune it\u2026 because Indian languages are natively supported.\u201d\u00a0<\/p> The government-backed consortium is betting that smaller, specialised models can deliver stronger real-world impact in emerging markets than massive general-purpose systems.<\/p> \u201cA majority of business, government and many consumer cases actually don\u2019t need a trillion-parameter frontier model,\u201d Bal said. A trillion-parameter system is \u201clike a thousand PhDs all bottled into one,\u201d he added, but most organisations instead need domain-specific expertise tailored to local conditions.\u00a0<\/p> India-first architecture<\/strong><\/p> Unlike many global models trained primarily on English-language data, Param 2 was built from the ground up for multilingual Indian usage.<\/p> BharatGen plans a rolling release strategy, with additional checkpoints and fine-tuned versions to follow. The model will also serve as a base layer for voice and vision systems.<\/p> \u201cWe have chosen to build single-modality multilingual models,\u201d Bal said. Param 2 will be \u201cvery text-focused\u201d initially, with separate releases planned for voice and document-understanding capabilities.\u00a0<\/p> Heavy compute investment<\/strong><\/p> Training Param 2 required substantial government-backed computing resources. BharatGen received one of the largest allocations under India AI Mission, including thousands of graphics processing units.<\/p> \u201cThe GPUs that we\u2019ve used so far from the AI Mission have gone into training of Param 2,\u201d Bal said, noting that the project has been using roughly 4,000 GPUs for over a month.\u00a0<\/p> The company plans to scale further, with larger models already in development.<\/p> \u201cYou should expect to see multiple size models\u2026 some small ones for edge inference\u2026 bigger ones like the 17B\u2026 and progressively larger models over the course of this year,\u201d he said.<\/p> <\/div> BharatGen, India AI Impact Summit 2026, OpenAI, Google, artificial intelligence, AI, India AI Mission, Param 2 #039Most #realworld #cases #dont #trillionparameter #AI039 #BharatGen #CEO #Rishi #Bal #Param #launch1771271676<\/p> ","protected":false},"excerpt":{"rendered":" BharatGen on February 16 launched Param 2, a 17-billion-parameter artificial intelligence model designed to serve India\u2019s linguistic and sectoral needs, positioning it as a domestic alternative to global systems from companies such as OpenAI and Google. The model, unveiled at the India AI Impact Summit 2026, supports 22 Indian languages. \u201cParam 2 launches today, that\u2019s […]<\/p>\n","protected":false},"author":1,"featured_media":3631,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[37],"tags":[12347,99,12350,2695,12352,12345,12348,735,458,75,10915,5810,217,216,12353,12346,4343,12351,12349],"class_list":["post-3630","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-content-marketing","tag-039most","tag-ai","tag-ai039","tag-artificial-intelligence","tag-bal","tag-bharatgen","tag-cases","tag-ceo","tag-dont","tag-google","tag-india-ai-impact-summit-2026","tag-india-ai-mission","tag-launch","tag-openai","tag-param","tag-param-2","tag-realworld","tag-rishi","tag-trillionparameter"],"acf":[],"_links":{"self":[{"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/posts\/3630","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3630"}],"version-history":[{"count":0,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/posts\/3630\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=\/wp\/v2\/media\/3631"}],"wp:attachment":[{"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3630"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3630"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/longzhuplatform.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3630"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}