NVIDIA-Backed Simplismart Raises $14 Mn Series B as AI Inference Demand Grows

Bengaluru-based Simplismart has raised $14 million, or around Rs. 133 crore, in a Series B funding round as more companies look for cheaper and faster ways to run AI models in real business applications. Earlier also, Nvidia was in advanced discussions to lead a $20 million funding round in Simplismart, with the startup being valued at around $100 million

Dallas Venture Capital led the round. NVIDIA, Accel, Shastra VC, Micromax, and angel investors also participated. The funding puts Simplismart in a growing group of Indian AI infrastructure startups working behind the scenes of chatbots, voice tools, video apps, and enterprise software.

What Simplismart does

Simplismart was founded in 2022 by Amritanshu Jain and Devansh Ghatak.

The company helps businesses deploy, run, monitor, and optimize open-weight AI models.

An open-weight model is an AI model whose core parameters are available for companies to use and adapt. This gives businesses more control over where their AI system runs, how it handles data, and how much it costs.

The difficult part is not always building an AI model. It is running that model reliably once thousands or millions of requests begin arriving. A voice assistant needs a quick response, while an overnight document-processing task can accept a slower response if it reduces the bill.

Simplismart’s software helps companies make those trade-offs. It works on the layer between an AI model and the GPU hardware used to run it.

Why the $14 Mn round matters

The Series B will support development of Simplismart’s inference engine and orchestration platform as per its Press release. Inference means the stage when a trained AI model receives a request and produces an answer, image, voice response, or other output.

The startup also plans to expand partnerships with cloud providers and enterprises, with North America among its growth priorities. Simplismart does not build or own a fleet of GPUs. Instead, it provides software that lets cloud providers turn their existing GPU capacity into managed AI services under their own brand.

For example, Yotta uses Simplismart to power the serverless inference layer of Shakti Studio. This allows customers to run AI models without having to manage all the underlying infrastructure themselves.

Simplismart says companies including HeyGen, Higgsfield, InVideo, Shiprocket, Apna, and Ema use its platform for video, voice, multimodal, and enterprise AI workloads.

NVIDIA’s role in the deal

NVIDIA’s participation is important because its GPUs are widely used to train and run AI models. Simplismart says its engineers work with NVIDIA Solutions Architects and use that technical collaboration to improve deployment and scaling for customers.

For Simplismart, NVIDIA is more than a financial backer. The company can also provide technical knowledge and industry connections as Simplismart works with cloud providers and large enterprises.

The startup said its annualized revenue run rate increased 14 times over the past six months. That is a company claim, but it suggests strong demand for infrastructure that can make AI applications more economical in production.

What about the reported $40-50 Mn round

Earlier reports said Simplismart was in discussions for a much larger funding round of around $40 million to $50 million, potentially at a higher valuation.

However, Simplismart’s chief executive Amritanshu Jain has clarified that the newly announced $14 million Series B is a separate, closed round. He also said the company is not actively raising money at present.

This means the larger round should be treated as earlier reported fundraising discussions, not as a confirmed future fundraise.

Competitors and market position

Simplismart competes in a crowded AI infrastructure market that includes cloud platforms, model-serving tools, and companies such as Modal. Major cloud providers also offer their own AI deployment services.

Simplismart’s pitch is that businesses should not have to choose only between the lowest cost and the fastest response. Its platform aims to balance speed, cost, data location, and reliability for each workload.