Bengaluru inference startup Simplismart raises $14 million, with Nvidia as a strategic investor

Bengaluru's Simplismart has raised $14 million in a Series B round led by Dallas Venture Capital, with Nvidia joining as a strategic investor.¹² Accel, Shastra VC and Micromax also took part.¹² The Hans India, citing a TechCrunch report, puts the company's post-money valuation at about $100 million.¹

What Simplismart sells

The company makes software that sits between AI models and the hardware they run on, tuning deployment settings so models answer faster and serve more requests for less compute.¹² It also offers orchestration for deploying and scaling models, and lets cloud providers sell their spare GPU capacity as ready-made AI services.²

Customers can run it on Simplismart's platform or inside their own cloud or on-premises setup.¹ Named customers include InVideo, HeyGen, Higgsfield and Ema.¹ Yotta uses Simplismart's technology for the serverless inference service on its Shakti Studio.¹

The numbers it shares

Simplismart claims its software cut InVideo's GPU serving costs by 56%.¹ It says its annualised revenue run rate grew 14 times over the past six months, and that more than half its revenue comes from the US even though most of its customers are in India.¹ These figures come from the company and have not been independently checked.

Chief executive Amritanshu Jain, a former Oracle engineer, founded the company in 2022 with Devansh Ghatak, who previously worked at Google.¹

Why Nvidia's name matters

A chip maker backing a company whose pitch is "use fewer GPUs" sounds odd at first. But inference is where AI spending is heading, and software that makes each GPU earn more helps the clouds that buy Nvidia's chips. For Simplismart, a strategic investor of that size helps when selling to cloud providers.

For Indian AI startups, the more useful signal is the revenue split. A Bengaluru company making most of its money from US customers shows that infrastructure software built in India can sell abroad, while Yotta's use of it shows a role in India's own compute build-out.

The money will go into the inference engine and orchestration platform, more cloud partners, and growth in North America and other markets.²

What to watch

• Whether more Indian GPU cloud providers adopt Simplismart for their inference services.
• Independent results on the cost savings it claims.

Related: the IndiaAI tender for 10,000 compute units and DeepSeek's open-source inference engine.