India Bets Big On AI: How It Plans To Be AI Superpower by 2030 | CNA

[冖_冖]

Indian companies look to Chinese LLMs as AI costs bite

DeepSeek and others are winning on price, but concerns over foreign dependence loom.

kr-asia.com

​

Indian companies look to Chinese LLMs as AI costs bite​

Written by Nikkei Asia Published on 21 Jul 2026

Indian companies are increasingly leaning on Chinese large language models (LLMs) developed by DeepSeek, Alibaba, and Moonshot AI to contain their artificial intelligence spends, extending India’s reliance on China for cutting-edge technologies despite a long history of standoffs between the neighbors.

Puneet Kumar, CEO at Mirae Asset Venture Investments India, said that several consumer technology startups that he has met since mid-2025 use such Chinese open-weight LLMs—those that rely on publicly accessible parameters—which help drive down costs by an “order of magnitude.”

Because these parameters are publicly available, users can download and modify them on their own computers, in contrast to the proprietary offerings by US frontier labs such as OpenAI and Anthropic. The Chinese open-weight LLMs can be accessed in India through service providers such as Microsoft at a fraction of the price of their American counterparts, thanks to their low cost of development.

“The US models are expensive, and for a lot of basic things, you don’t need them,” Gupta said. “It’s overkill, like trying to drive a sports car on a crowded city road.”

For the DeepSeek models that Microsoft makes available in southern India through its Foundry platform, charges range between USD 0.19–1.74 per million input tokens—the tiny units of data that AI models process—while the prices for output tokens range between USD 0.51–5.40 per million. Input costs for Moonshot’s Kimi go up to USD 0.95, while output costs can reach USD 4 per million tokens.

Companies like Coinbase, DoorDash, and Airbnb have publicly said that they have begun using Chinese models.

Vidya Madhavan, founder of Elevation Capital-backed dating app Schmooze, said that possibilities of “substantial” savings, coupled with wider uptake of open-weight LLMs globally and their ability to deliver a satisfactory performance in comparison with their US peers, encouraged her to deploy Alibaba’s Qwen models after some initial hesitation.

 

[冖_冖]

Chinese Qwen AI Overtakes Meta, Google for Indian BFSI

Chinese open-weight AI model Qwen now leads Meta and Google in download volumes, becoming the default choice for new-age Indian BFSI players. Fintechs like Inno

catchthebrief.com

Chinese Qwen AI Overtakes Meta, Google for Indian BFSI​

Chinese open-weight AI model Qwen now leads Meta and Google in download volumes, becoming the default choice for new-age Indian BFSI players. Fintechs like Innoviti are rapidly adopting it for cost savings and RBI compliance, testing models and deploying in just three days. Legacy financial institutions, however, struggle with the tech chops and inherent distrust of Chinese models.

How We Got Here

Qwen actually overtook Meta and Google's AI models in download volumes in August, just before Innoviti upgraded to its latest 3.8 version in early September. This open-weight approach makes model "weights" public, unlike open-source models, while keeping source code and training data proprietary.
The Numbers

• Innoviti tested Qwen 3.8 for photo verification of payment terminals for just three days, achieving near-zero false positives and less than 5% false negatives.
• Innoviti’s use case evolved from multimodal image processing to core software engineering tasks like code generation and testing with the new Qwen version.
• Stock brokerage Zerodha uses Qwen, with its engineers self-hosting the model directly on personal laptops for programming work, ensuring no customer data resides there.
• Open-weight models offer public weights but shield proprietary training information, allowing users to run them on private servers for greater data control.
• These models are cheaper and more customisable than closed-source alternatives, providing greater visibility and auditability into their internal workings.

What Happens Next

The market will closely watch how quickly Indian AI makers can build trust and fill the existing gap, especially with legacy players who might face increasing pressure to modernize their stack by H2 2024. A key indicator will be the accelerated adoption rate within non-banking financial companies (NBFCs).

Why This Matters for India

For founders building AI solutions in Bangalore and Hyderabad, this signals a massive opportunity to white-label Chinese models, offering the cost and control advantages that current Indian-made options struggle to match for financial services.

The Take

The winners here are clearly the deep-tech founders who understand how to integrate and secure these "bring-your-own-model" stacks. Legacy banks will continue to lose ground to agile fintechs, unless they acquire a capable AI services firm by early 2025.