Simplismart Raises $14 Million Series B With NVIDIA and Dallas Venture Capital

Bengaluru-based Simplismart has raised $14 million in a Series B round led by Dallas Venture Capital, with NVIDIA, Accel, Shastra VC and Micromax participating. The company, which builds the software layer that sits between AI models and GPU infrastructure, said its annualized revenue run rate grew 14-fold over the past six months.

TechCrunch described the deal as an all-primary round. Simplismart will use the money to expand its AI inference platform, deepen cloud partnerships and grow across North America and other global markets.

What does Simplismart actually sell?

Simplismart builds the software layer that turns third-party GPU fleets into production AI infrastructure. Its modular inference engine lets businesses configure how AI models run, trading off speed, cost and performance. Its orchestration platform manages deployment and scaling across cloud and on-premises setups.

The model matters less than the execution, according to CEO Amritanshu Jain. "The advantage comes from how you run them: what each interaction costs, how quickly the application responds, and whether it stays reliable as demand grows," Jain said.

Jain, co-founder and chief executive, framed the market position bluntly in comments to TechCrunch: "Nobody actually needs the fastest or the cheapest inference." Businesses instead need AI models tuned to their own applications — a specific response speed, a specific request volume, a specific compute budget. Simplismart said its platform optimizes all three, and lets companies deploy through its platform or inside their existing cloud and on-premises infrastructure.

Who uses it, and what results do they see?

The customer list includes AI video-generation platform InVideo, HeyGen, Higgsfield, Shiprocket, Apna and Ema. Simplismart said these customers run production workloads across video, voice, multimodal and enterprise AI applications.

The most concrete result the company cited: it helped InVideo cut GPU serving costs by 56%. Simplismart also works with Yotta, an Indian cloud provider, and with enterprise AI startup Ema.

Yotta uses Simplismart's technology for the serverless AI inference service inside its Shakti Studio platform. Simplismart positions this as a playbook for cloud providers: they can offer managed AI services under their own brands without building their own software stack from scratch.

The company draws a sharp line between itself and inference providers that buy or reserve GPU capacity and resell it. Simplismart sells software, not compute.

Why did NVIDIA invest?

NVIDIA participated in the round and worked with Simplismart on technical integrations around its Dynamo inference software. Simplismart engineers and NVIDIA solutions architects collaborated on that effort, according to TechCrunch.

For NVIDIA, the investment extends its reach deeper into the efficiency layer of AI deployment. Dayakar Puskoor, managing partner at Dallas Venture Capital, made the economic case: "Inferencing is emerging as one of the biggest game changers in AI economics." Dallas Venture Capital linked its investment to helping enterprises run AI workloads more efficiently.

Where does the money go next?

The funding will support expansion of cloud-provider and sovereign AI partnerships, plus growth in North America and other global markets.

The technical roadmap points at a specific bottleneck. Co-founder and CTO Devansh Ghatak said the funding will back work toward "a dynamic GPU scheduling layer that can match AI workloads with available computing capacity across regions and time zones." In plain terms: GPUs sit idle in one region while workloads queue in another, and a scheduling layer that moves compute demand to supply would raise utilization across the board.

The Economic Times separately reported CEO Jain confirming the 14x growth in annualized revenue run rate over the past six months.

The bet from Dallas Venture Capital and NVIDIA is that inference economics, not model training, becomes the decisive cost line for enterprise AI — and that the company selling the scheduling and orchestration software captures more value than the one reselling raw GPU hours.