NxtGen Launches M: An Open-Source GenAI Platform Built in India

NxtGen, which is positioning itself as a strategic player in the sovereign cloud and AI infrastructure space, has announced the general availability of “M,” a next-generation open GenAI platform that transforms human intent into real-world outcomes using intelligent, agentic workflows. This is the company’s homegrown generative AI platform designed to bridge the gap between cutting-edge AI capabilities and real-world, everyday use cases.

Why the name “M”? The company said it reflects the first sound many humans ever make, “M” symbolizes the beginning of awareness, intelligence, and connection. It pays homage to human origin while being engineered to power the future of digital interaction.

Speaking at the media briefing, Rajgopal A.S, NxtGen MD & CEO, compared AI’s transformative potential to electricity, noting that while AI innovation is currently concentrated in a few hands globally, platforms like M can democratise access.

Positioned as an open-source-driven, sovereign-ready alternative to closed, global models, M aims to empower enterprises, startups, and the public with AI tools that integrate directly into daily workflows and services.

Talking about the inspiration behind M, Rajgopal said, “We asked ourselves — can AI be a public utility? The reality today is that it’s already getting concentrated in a few hands. Our goal with M is to break that mould, leveraging open-source to create something that works for everybody.”

The Vision Behind M

Rajgopal underscored India’s dual advantage in AI: a strong services ecosystem and an emerging policy framework for AI innovation. However, he warned of infrastructure gaps that could leave Indian enterprises and startups behind without intervention.

“India has the talent, but we need the infrastructure and accessible platforms. Without them, the general public, enterprises, and especially startups will struggle because they lack the tools and talent to compete at scale,” he said.

NxtGen has built M as an AI orchestration layer that can work with thousands of open-source large language models and AI components available in the global developer ecosystem. It can select the most relevant one for a given task and integrate with external systems to execute actions — not just generate text.

This development builds on NxtGen’s broader AI infrastructure plans. The company has publicly announced significant investments to expand its GPU capacity and is ramping up its AI engineering workforce to meet growing demand.

How M Works

M uses a “smart model” middleware that dynamically decides which AI model to use based on the user’s request. For domain-specific tasks, it can switch to specialised models trained on Indian financial, legal, or other sector-specific datasets.

At its core, M uses an open-source model to analyse what the user is expecting and dynamically invokes the most appropriate open-source model for each task. This multi-model approach is transparent to the user, enabling seamless interaction across a wide range of services while preserving context, accuracy, and speed.

M leverages open source as a base and is architected to always use the best available models. It leverages Llama 4 models, the DeepSeek 671B model for expert reasoning and logic, and this is possible due to NxtGen’s advanced AI infrastructure. The company says: as India builds its own models and leverages AI for education, research, and increased productivity, M is the gateway for AI technology for the world to benefit from. M is ensuring that India is not losing time while we build for the world.

The platform supports text and speech inputs, understands APIs, and can connect to external services — from booking a movie ticket to scheduling a hospital appointment, as long as those services provide API access. “For instance, if two taxi providers open their APIs and one doesn’t, M will work with those that do. Over time, competitive pressure will likely force the others to participate,” Rajgopal noted.

The company also said that the design emphasises open-source sovereignty. For instance, the core models are open source and run on NxtGen’s infrastructure without retaining user data. Internet access, when required, is handled through a controlled process outside the core model, reducing security risks.

“Most of the internet already runs on open source. With M, there’s no closed-source dependency for scale,” Rajgopal said. “We don’t let the model extract external data directly. The architecture has guardrails.”

Live Demonstrations: From Code to Live Outcomes

At the launch, NxtGen showcased several live demos to illustrate M’s capabilities:

Rapid Website Development — In a live session, M researched India’s economic trends, generated a business plan, built a functional website with design elements, and even produced a logo — all without pre-scripting.

Voice-Driven Actions — Using only speech, a user instructed M to provision a virtual machine on NxtGen’s Speed Cloud platform, demonstrating integration with enterprise infrastructure.

Document Intelligence — M can process PDFs, spreadsheets, and other documents, extracting and reasoning over their contents to answer queries in context.

Rajgopal stressed that these were not lab experiments but deployable features. “We’ve already built numerous enterprise use cases. The challenge for startups is not technology but market adoption. M integrates their use cases into a platform the public and businesses can access immediately.”

Bridging Enterprise and Public Use

NxtGen envisions M as both a B2B enabler and a public-facing AI utility. Enterprises can integrate their services via APIs, while public-facing integrations could remove friction from common citizen services such as passport or voter ID applications, train ticket booking, or accessing government schemes.

Rajgopal is clear that some of the most impactful use cases could come from public services: “I don’t want citizens to go through 50 questions to find a polling booth location. With M, you could just ask and get the answer instantly, over voice.”

How M is Competing With Global Models

When asked how M stacks up against closed-source leaders like OpenAI or Anthropic, Rajgopal emphasised flexibility and openness over proprietary lock-in.

“If tomorrow there’s a better model, we replace it. We haven’t sunk costs into building a single foundation model; we’ve built the infrastructure to integrate the best available models for each task.”

The platform’s ability to maintain conversational context over long interactions, without restricting token limits, allows it to handle complex, multi-step reasoning better than many frontier models, he claimed. “The qualitative difference shows when you give a detailed, paragraph-length query. The depth of response is higher because we don’t limit the model.”

Addressing enterprise concerns about security, Rajgopal noted that NxtGen’s architecture prevents the model from directly accessing the internet or external systems without controlled mediation. This makes M more suitable for sovereign and regulated workloads.

The Road Ahead for M

NxtGen is inviting startups, SMEs, and enterprises to integrate their services with M, with Rajgopal urging feedback to improve the platform. The company aims to rapidly expand its library of integrated use cases, with adoption driven by utility rather than marketing.

Future plans include deeper vertical specialisation — from financial compliance and legal advisory to healthcare automation — and more real-world agency capabilities, such as scheduling and executing tasks autonomously when triggered by natural language instructions.

Rajgopal’s closing message was clear: “We want to take AI to everybody, drive real-world use cases, and make it easy. M is built for the world, but its thought process comes from India.”

Also Read

NxtGen’s Bold Bet: India’s Sovereign, Enterprise-Ready AI Stack