On a date not specified, Nvidia participated in a $14 million Series B financing for Indian AI startup Simplismart, a move aimed at supporting software that can run AI models more efficiently and sometimes with fewer GPUs.
The round, led by Dallas Venture Capital, also included Accel, Shastra VC and Micromax, and placed the four-year-old company at an estimated $100 million valuation after the new capital.
Enterprises are spending heavily on GPUs for AI workloads, yet achieving reliable and cost-effective scaling remains difficult. Different tasks require trade-offs among speed, cost and performance, and deploying models often involves complex configuration. Simplismart claims its platform automates much of this, adjusting model deployment on existing infrastructure so customers need not manage those settings themselves.
"Nobody actually needs the fastest or the cheapest inference," said Amritanshu Jain, co-founder and CEO of Simplismart, in an interview. He noted that firms prefer AI speeds and costs that align with their specific use cases.
The Bengaluru-based firm positions its software between AI models and the hardware they run on, tweaking deployment and management to meet client needs. It says it can fine-tune response times, request volumes and power consumption, allowing firms to extract more value from their current GPU assets.
Clients can run models via Simplismart's own platform or integrate it with their cloud or on-premise setups. The suite also offers model fine-tuning, automatic scaling with demand shifts, and performance monitoring. Cloud providers can use the technology to build AI inference services on GPU infrastructure.
Simplismart reports that AI video-generation platform InVideo reduced its GPU serving costs by 56% using the software. Other customers include video AI firms HeyGen and Higgsfield, enterprise AI startup Ema, and Indian cloud provider Yotta, which powers a serverless AI inference service in its Shakti Studio platform.
According to Jain, the startup's annualised revenue run-rate grew fourteen-fold over the past six months to an eight-figure amount, based on recent usage and excluding pass-through infrastructure revenue; the exact figure was not disclosed.
More than half of the revenue originates from the United States, even though the company has a larger customer base in India, Jain said.
Beyond funding, Nvidia is collaborating with Simplismart on technical integrations, such as around Nvidia's Dynamo inference software, involving both companies' engineers and solution architects, and Nvidia has helped introduce the startup to potential partners.
Although Nvidia dominates AI chip supply, its backing of software that may reduce GPU counts appears paradoxical. Jain argued that improving inference efficiency ultimately benefits Nvidia by making its hardware more appealing and expanding AI adoption.
Simplismart operates in a crowded AI inference software market, competing with Nvidia's own offerings and major cloud providers like Amazon Web Services, Google Cloud and Microsoft Azure, as well as numerous startups developing tools to speed up and cheapen AI model execution.
"Where we tend to win is where people have to be more stringent on their SLAs [service-level agreements] and TCO [total cost of ownership]," Jain explained, emphasizing the firm's focus on customized balance rather than preset fast or cheap options.
Founded in 2022 by former Oracle engineer Jain and former Google engineer Devansh Ghatak, who met at BITS Pilani, Simplismart now employs 58 people.
The fresh capital will be used to grow the U.S. presence, expand the team and further develop the inference platform, with the United States remaining a key market.
Earlier reports in The Economic Times suggested Simplismart was negotiating a $40 million to $50 million raise at a $300 million to $400 million valuation. Jain declined to comment, stating the startup was not actively fundraising and that the Series B round was closed and would not be extended.
