As companies race to deploy artificial intelligence, the cost of running AI models, known as inference, has become one of the biggest bills in technology. A Bengaluru startup that helps companies cut that cost has won backing from the world's most valuable chipmaker.
Simplismart has raised $14 million in a Series B round led by Dallas Venture Capital, with participation from Nvidia, Accel, Shastra VC and Micromax, according to TechCrunch and StartupTalky. The all-primary round values the four-year-old company at about $100 million, including the new capital.
What Simplismart does
Simplismart's software sits between AI models and the computing hardware that runs them. It tunes how models are deployed to balance response speed, the number of requests that can be handled, and the amount of computing power used.
Customers can run the software on Simplismart's own platform or within their own cloud or on-premises infrastructure. The company also offers fine-tuning, autoscaling and performance monitoring, and allows cloud providers to run their own inference services using its technology.
"Nobody actually needs the fastest or the cheapest inference," co-founder and chief executive Amritanshu Jain told TechCrunch. The company's approach is to tailor the trade-off between performance and cost to each customer's needs, rather than offering fixed options.
Real savings for customers
Video creation platform InVideo reportedly cut its GPU serving costs by 56% using Simplismart, according to TechCrunch. Other customers include AI video company HeyGen, Higgsfield and Ema. Indian cloud provider Yotta uses Simplismart's technology to power a serverless inference service.
For AI companies, GPU costs can account for a large share of their expenses. Reducing the number of GPUs needed to serve the same number of users can significantly improve their economics.
Rapid growth
Simplismart's annualised revenue run rate grew 14-fold over six months to an eight-figure dollar sum, according to TechCrunch. The figure reflects recent usage rather than contracted recurring revenue and excludes pass-through infrastructure costs. More than half of revenue comes from the United States, even though the company has more customers in India.
Nvidia's involvement
Nvidia is working with Simplismart on technical integrations, including with Dynamo, Nvidia's inference software, and has helped introduce the startup to potential customers and partners. Nvidia has invested in a growing number of AI startups worldwide, including in India, as it seeks to expand the ecosystem around its chips.
Interestingly, Simplismart also competes with Nvidia, as well as with Amazon Web Services, Google Cloud and Microsoft Azure, all of which offer inference services. "Where we tend to win is where people have to be more stringent on their SLAs," Jain said, referring to service-level agreements that guarantee performance.
The founders
Simplismart was founded in 2022 by Amritanshu Jain, a former Oracle engineer, and Devansh Ghatak, a former Google engineer, who met at BITS Pilani. The company has 58 employees.
Where the money goes
Simplismart plans to use the funding to expand its presence in the United States, grow its team and improve its software. The company previously raised a $7 million Series A led by Accel.
The Economic Times had reported that the company was in talks to raise $40 million to $50 million at a valuation of $300 million to $400 million. Jain declined to comment on that report, telling TechCrunch that the company is not actively fundraising and that the Series B is closed and will not be extended.
Why inference matters
The AI industry's focus is shifting from training models to running them. Training a large model is a one-time, if enormous, expense. Inference happens every time a user asks a question, generates an image or uses an AI feature, and scales with usage. As AI adoption grows, inference costs can come to dominate total AI spending.
That shift has created opportunities for companies that make inference more efficient, through software optimisation, specialised chips or better infrastructure. Startups such as Simplismart are betting that many companies will need help running AI models efficiently, particularly as they deploy open-source models rather than relying entirely on proprietary services.
India's AI infrastructure ambitions
Simplismart is part of a growing group of Indian startups building AI infrastructure, rather than only applications. The government's IndiaAI Mission has funded GPU capacity for Indian startups and researchers, and companies such as Yotta have built large GPU clusters. Efficient inference software can help make the most of that capacity.
The fact that more than half of Simplismart's revenue comes from the US shows how Indian deep-tech startups can compete globally in technically demanding markets.
The economics of inference
As AI moves from experimentation to production, the cost of inference, running trained models to produce answers, images or video, is becoming a major concern for businesses. Training a large model is a one-time expense, but inference costs recur every time a user interacts with an AI application, and they scale with usage. For consumer applications with millions of users, inference can become the largest component of operating costs.
Companies such as Simplismart aim to reduce those costs by optimising how models run on hardware, using techniques such as better scheduling, model compression and efficient use of graphics processing units. The reported 56% reduction in GPU costs for InVideo illustrates the potential savings. For startups building AI products, such savings can make the difference between a viable business and one that loses money on every customer.
Indian AI startups attracting global capital
Simplismart's funding adds to a growing list of Indian AI startups that have attracted investment from global technology companies and venture firms. Nvidia's participation is a signal of confidence, given its central role in the AI industry and its interest in an ecosystem of companies that make its hardware more efficient and attractive to customers. Indian AI infrastructure startups are increasingly building products for global markets from the outset.
What comes next
Simplismart will focus on expanding in the US and deepening its integration with Nvidia's software. Its growth will be watched as a test of whether a specialised inference company can thrive alongside the hyperscale cloud providers, and of whether Indian AI infrastructure startups can build globally competitive businesses. Customers and investors alike will be watching whether Simplismart can sustain its rapid revenue growth as competition in inference optimisation intensifies, with larger cloud providers and specialist startups all targeting the same opportunity.
