Nvidia backs Simplismart's $14M Series B to optimize AI inference efficiency
Nvidia joins $14M funding round for Indian startup Simplismart, which helps companies reduce GPU costs and run AI models more efficiently.
Last verified:
Nvidia has joined a $14M Series B funding round for Simplismart, a Bengaluru-based startup that optimizes how AI models consume GPU resources, according to TechCrunch AI. The all-primary round, led by Dallas Venture Capital and including participation from Accel, Shastra VC, and Micromax, values the four-year-old company at approximately $100M. The investment reflects a strategic bet by Nvidia on software abstraction layers that can reduce GPU dependency—a counterintuitive move by a chip vendor seeking to maximize hardware sales.
Simplismart’s inference-optimization approach
Simplismart operates as a middleware layer between deployed AI models and underlying GPU infrastructure. According to TechCrunch AI, the software automates tuning of deployment parameters—including response latency, request concurrency, and power draw—without requiring manual configuration by customers. Rather than selling the fastest or cheapest inference, CEO Amritanshu Jain told TechCrunch that the platform delivers inference economics tailored to each application’s actual business constraints.
The startup’s customer roster spans video-generation platforms (InVideo, HeyGen, Higgsfield), enterprise AI companies (Ema), and cloud providers (Yotta, which integrates Simplismart into its serverless inference service). Simplismart’s most cited impact comes from reducing InVideo’s GPU serving costs by 56%, a figure that illustrates the cost-arbitrage opportunity in inference optimization.
Growth trajectory and revenue composition
TechCrunch AI reports that Simplismart’s annualized revenue run rate has expanded 14-fold over the preceding six months, reaching an eight-figure threshold without exact disclosure. The figure reflects usage-based metrics rather than contracted annual recurring revenue (ARR) and excludes pass-through revenue to infrastructure providers. Notably, more than half of revenue originates from the US market despite the majority of paying customers residing in India—a geographic split suggesting higher deployment density or contract values in North America.
Why this matters
Nvidia’s participation in the round signals the chipmaker’s acknowledgment that GPU utilization efficiency—not raw capacity—has become a competitive differentiator for cloud operators and enterprises facing sustained pressure on AI infrastructure budgets. As GPU scarcity eases and pricing normalizes, software-defined optimization will likely determine which cloud providers and enterprises capture margin in inference workloads. Startups like Simplismart that abstract away hardware-specific tuning may accelerate adoption of AI services among price-sensitive segments, while also making the case that additional GPU procurement is not always the solution to scaling inference economically.
Frequently Asked Questions
What does Simplismart's software do?
Simplismart's platform sits between AI models and GPU infrastructure, automatically optimizing deployment parameters like inference latency, request throughput, and power consumption to meet specific business requirements without requiring additional hardware.
Who is investing in Simplismart?
Nvidia, Accel, Shastra VC, and Micromax participated alongside lead investor Dallas Venture Capital in the Series B round.
What is Simplismart's revenue run rate?
According to CEO Amritanshu Jain, the startup's annualized revenue run rate has grown 14-fold in the past six months to an eight-figure sum, though the exact figure remains undisclosed.