Open source in the age of AI: Powering India’s next innovation wave at TechSparks 2025

India’s tech community has long thrived on open source, from Drupal and Free and Open Source Software (FOSS) communities to Linux adoption in startups and enterprises. 

Today, with AI reshaping industries, open source is accelerating AI-driven innovation. 

OpenAI’s closed-weight models once defined breakthroughs, but now models such as Meta’s LLaMA, Google’s Gemini, Alibaba’s Qwen, DeepSeek‑R1, IBM Granite, and Databricks’ DBRX are open source or open weight, and are accessible to Indian startups, developers, and enterprises.

In fact, in tech stacks across sectors, such as fintech, healthtech, edtech, and ecommerce, open source models are becoming default building blocks. 

Developers in India are now customising, deploying and scaling them locally to build AI‑enabled apps that reflect real user needs, not just generic chatbot templates.

In line with this, the upcoming edition of TechSparks, YourStory’s flagship tech-startup summit, will focus on highlighting the trend of open-source tech in the age of AI and uncover its current and upcoming applications across various sectors.

TechSparks 2025, with the theme ‘India 2030: Powered by AI’, will take place in Bengaluru  from November 6 to 8, bringing together ecosystem leaders leading the world in open source and AI.

Why open source matters now

A major example of Indian innovation in open source and AI is BharatML Stack from Meesho, a multi-lingual open LLM infrastructure built to serve Indian languages and grassroots innovation.

"We put BharatMLStack to the test during high-traffic events like our Mega Blockbuster Sale in March 2025, where it delivered at scale, demonstrating its ability to perform under peak load conditions," said Sanjeev Kumar, Founder and CTO of Meesho, in a statement at the time of launch.. 

He added, "This helped Meesho drive higher user engagement, better conversions, and increased order volumes during the sale. By open sourcing it, we’re sharing a high-scale, AI stack with the broader tech community, purpose-built for real-time use cases and tailored for Indian businesses.”

According to a McKinsey‑Mozilla survey of over 700 global AI leaders, more than 50% of enterprises use open-source AI tools across data pipelines, models, and tooling, especially when they view AI as central to competitive advantage 

Respondents cited open-source tools as 60% cheaper to implement and 46% cheaper to maintain than proprietary tools, though the latter offer faster time‑to‑value.

Developers also value the open-source experience: 81% say it’s essential for job satisfaction, while 66% believe it adds tangible professional value.

Leveraging this are Indian companies that have begun building their own AI systems that work with open source LLMs and are trained on relevant data,

Fractal’s Fathom‑R1 14B, a reasoning model, launched as open-source tech, is a recent example.

“Today’s large pre-trained AI models are great at summarisation, information retrieval, and content generation. But the next frontier is building systems that can think, reason, plan and act. It’s about building systems that can work with pre-trained models - and accomplish complex real-world tasks through better reasoning,” said Srikanth Velamakanni, Co-founder of Fractal, in a LinkedIn post when the model was launched..

“That’s why we, at Fractal, proposed building India’s first Large Reasoning Model (LRM), a next-generation AI system built to work with open source LLMs, trained on Indian data, and designed to tackle real-world complexity. It’s a bold step, but a necessary one,” he added.  

Open‑source AI in India’s innovation engine

As per the Linux Foundation, flexible licensing and community ecosystems help reduce vendor lock‑in and promote trust.

An example of this in India’s tech ecosystem is Juspay’s open‑sourcing of its payment orchestrator in fintech, with the aim to create developer interest and spark startups to reuse its architecture in new financial flows and in an open manner. 

The goal, according to Juspay, is to eliminate vendor lock-in and empower merchants with a fully customisable and compliant payment system.

“Our vision is to make payments open, transparent, and accessible so that businesses can focus on growth rather than payment complexities,” said Sheetal Lalwani, Co-founder and COO of Juspay. 

“With the launch of the open source routing engine, we are making payments orchestration a truly interoperable and community-driven software.”

These efforts reveal a shift: open source is no longer just code sharing. It’s a platform for scaling innovation and building enterprise‑grade IP from India outward.

(Learn more on how AI is reshaping India’s startup ecosystem and be part of this change only at TechSparks 2025. Join us at Taj Yeshwantpur, Bengaluru, on November 6–8 and be part of the innovation shaping the nation’s future. For more information, click here.)