Frames and Fans
Live
Builds

India targets homegrown GPU within two years

India's government has set a 24-month deadline to develop an indigenous graphics processing unit under its Semiconductor Mission 2.0, aiming to reduce reliance on foreign chip suppliers and capitalize on the AI boom.

India's government has set a 24-month deadline to develop an indigenous graphics processing unit under its Semiconductor...

India plans to create a homegrown GPU within 24 months. S. Krishnan, secretary of the Ministry of Electronics and Information Technology (MeitY), announced the target on November 4 as part of the India Semiconductor Mission 2.0. The initiative aims to develop domestic AI accelerator chips. The GPU project is at an early planning stage. No manufacturer, chip supplier, or performance targets have been disclosed.

India's semiconductor market exceeded $50 billion and is expected to approach $100 billion by 2030, according to the India Electronics and Semiconductor Association. The government aims for local companies to capture more of this growth. The global GPU market is dominated by a few vendors. Chip development typically takes years due to advanced process technology and EDA tools. This makes India's 24-month goal highly ambitious.

The India Semiconductor Mission 2.0 (ISM 2.0) was approved by the Cabinet in July 2026. It has an allocation of roughly ₹1.27 lakh crore, approximately $15 billion, over 12 years. The mission explicitly includes AI-specific accelerators and a full semiconductor ecosystem. As of late 2026, five of the twelve projects from the earlier ISM 1.0 have entered commercial production.

Semiconductor ecosystem expands with new partnerships

Major partnerships are driving India's chip manufacturing growth. ASML has partnered with Tata Electronics to equip India's first commercial semiconductor fabrication facility in Dholera, Gujarat. The fab will use 300mm wafers with a planned capacity of 50,000 wafers per month.

Intel has announced a strategic partnership with the Tata group for semiconductor manufacturing and packaging in the Indian market. Several companies are manufacturing servers locally. VVDN, Kaynes, and Avalon are producing servers in India. C-DAC developed the Rudra server range for AI workloads and transfers that technology to these local manufacturers.

The Design Linked Incentive (DLI) scheme has supported 24 companies. These firms have secured venture capital or private investment. Their products include microcontrollers, drone flight controllers, smart meter chips, and CCTV cameras.

RISC-V processors power early-stage IoT applications

The proposed GPU is separate from existing SHAKTI and VEGA processors. These are CPU designs based on the open-source RISC-V architecture. SHAKTI and VEGA processors are advancing toward commercialization after years of development.

At the SEMICOM India event on September 26, the government showcased projects using these processors. They included an autonomous surface-cleaning robot, a digital telescope, and a forest monitoring system. The SHAKTI software development kit was released by IIT Madras in 2019. This enabled application development before the processor was commercialized.

In 2024, the Aries 3.0 development board launched. It features C-DAC's VEGA ET301 processor. The board is intended for simple applications like sensors, wearables, and IoT devices, not for desktop PCs or graphics-intensive workloads.

AI job transformation drives demand for local chips

AI adoption is expected to reshape India's workforce. The country's IT-enabled service sector employs about 6 million people. A NITI Aayog estimate suggests AI could make 2 million jobs redundant while creating 4 million new ones.

Krishnan compared the AI wave to the Y2K problem of 2000. He stated it could create a growth window for India's IT services industry. The GPU initiative is part of a five-layer framework for AI development. The layers include energy, accelerator chips, computing infrastructure, AI models, and applications.

Worldwide supply of AI GPUs has been tight. Prices for the chips have climbed. India accounts for about 20% of the global semiconductor design workforce. The government's target is to deliver a homegrown GPU product in about 24 months under Semiconductor Mission 2.0.

Topics

#Builds

Related coverage

More from Builds