Nvidia has invested in Bengaluru-based startup Simplismart as part of a $14 million Series B round, aiming to help companies run AI models more efficiently—sometimes with fewer GPUs. The round also involved Accel, Shastra VC, Micromax, and Dallas Venture Capital, with the valuation standing at about $100 million following the funding. Simplismart, founded in 2022, develops software that bridges AI models and hardware infrastructure to optimize how models run—balancing cost, speed, and performance automatically.
How Simplismart’s Software Works
Simplismart’s solution positions itself between machine-learning models and the systems they run on—whether in the cloud or on-premises. It handles deploying, managing, and tuning models to meet specific requirements for inference speed, throughput, and compute use. Customers can deploy via Simplismart’s platform or adapt it to their existing infrastructure. Features include automatic resource scaling, fine-tuning of models, and real-time performance monitoring. Cloud providers also leverage Simplismart’s tech to build their own AI inference services.
Real-World Impact and Market
Simplismart reports helping AI video-generation platform InVideo slash GPU serving expenses by 56% using its software. Other clients include HeyGen, Higgsfield, and enterprise AI company Ema. In India, cloud provider Yotta uses Simplismart’s module within its Shakti Studio service.
Revenue-wise, the startup has seen its annualized run rate jump 14-fold over the past six months, reaching into eight-figure territory. That figure reflects customer usage—not contractual commitments—and excludes amounts passed to infrastructure providers. While most clients are Indian, over half of revenue comes from the U.S., marking international traction.
Nvidia’s Role and Competitive Landscape
As part of the deal, Nvidia has collaborated on technical integrations with Simplismart—including work around Nvidia’s Dynamo inference software. Its engineers and solutions architects are involved in guiding deployment, and Nvidia is facilitating introductions to potential customers and partners.
On the surface, it might seem contradictory for Nvidia—primarily known for selling powerful GPUs—to back a company that helps reduce GPU usage. Simplismart argues that better efficiency doesn’t threaten Nvidia’s hardware business; instead, it can drive broader adoption by making deployment more cost-effective and practical.
Simplismart competes in a crowded space alongside Nvidia itself, as well as cloud giants like AWS, Google Cloud, and Microsoft Azure. Its strategy hinges on meeting strict performance goals (SLAs) and tight total cost of ownership constraints, delivering custom optimization rather than offering preset inference modes.
Founders, Team, and Future Plans
Co-founders Amritanshu Jain (formerly at Oracle) and Devansh Ghatak (formerly at Google), who met at BITS Pilani, lead the company which now has 58 employees. The fresh funds will support expansion in the U.S., growing the team, and further R&D into its inference platform. Simplismart confirmed the Series B round is closed and will not be extended, despite rumors of a larger raise at a higher valuation.
Simplismart’s rise illustrates urgency across the AI stack: as demand for models skyrockets, making them efficient becomes essential. Enterprises and infrastructure providers alike are chasing lower costs without compromising reliability or speed. Simplismart’s technology, bolstered by Nvidia’s investment and collaboration, may become a key enabler in scaling AI sustainably.
What to watch: whether Simplismart manages to hold its competitive edge as cloud providers build similar efficiency layers; how it differentiates its offerings amid heavy competition; and whether its U.S. expansion brings in more enterprise-scale contracts that solidify its market position.
