The AI race moves at full speed. Every big tech company wants more power for AI tools. This demand creates a huge problem. AI firms need many chips, but each chip system works in a different way. A shift from one chip to another takes time, money, and expert work.
A startup named Decart now wants to solve this issue. The company raised $300 million in a fresh funding round. Big names joined the deal, such as Nvidia, Radical Ventures, Benchmark, and Sequoia Capital. The new deal pushed Decart’s value close to $4 billion.
Why Decart Matters
AI companies depend on advanced chips to train models and answer user questions. Most firms use chips from Nvidia because Nvidia controls a large share of the AI chip market. Still, many companies now test other systems from Amazon and Google to lower costs and avoid shortages.
This move creates a challenge. AI models do not shift easily between chip systems. Each platform needs special software and custom code. A full transfer may cost millions of dollars.
Decart built software that removes much of this pain. The company created a system called Decart Optimization Stack, also known as DOS. This software helps AI developers run models on many chip types with less effort.
Dean Leitersdorf, one of the founders, said many firms sign contracts worth hundreds of millions just to move models between hardware systems. He called Decart “the great equalizer.”
Nvidia’s Surprise Investment
The most interesting part of the story came from Nvidia’s role in the funding round.
Decart helps companies use chips from Nvidia rivals such as Amazon Trainium and Google TPUs. At first glance, this setup may look bad for Nvidia. Still, Nvidia chose to invest in the startup.
This step shows how important Decart became in the AI market. Nvidia likely sees value in software that supports the whole AI ecosystem. If AI firms grow faster, demand for chips may rise across the industry.
The investment also shows Nvidia’s confidence. The company may believe its chips will stay strong even if customers gain more freedom to switch systems.
A Young Startup With Fast Growth
Decart started in 2023. Three Israeli engineers founded the company. The founders include brothers Dean and Orian Leitersdorf along with Moshe Shalev.
The brothers already built strong reputations in Israel’s tech sector. Dean earned a Ph.D. from Technion at age 23. Orian later broke that record after he finished his doctorate at age 21.
Despite its young age, Decart grew very fast. The startup raised $153 million in August 2025 before the latest funding round. Investors from both Silicon Valley and global tech circles rushed to support the company.
This rapid rise reflects one clear truth. Investors believe AI infrastructure may become one of the biggest business sectors of the decade.
Amazon Already Uses Decart Technology
Decart already works with large customers. Amazon became its biggest customer so far.
Amazon uses Decart technology across several business units. These include Twitch, online retail operations, and movie studios. Amazon also supplies Trainium chips, which compete with Nvidia hardware.
This partnership gives Decart strong market credibility. Big companies rarely trust core AI systems to small startups unless the technology offers real value.
The relationship also highlights a major trend. Tech giants now search for ways to cut dependence on Nvidia’s expensive chips.
More Than a Chip Software Company
Decart does not focus only on chip optimization. The startup also develops “world models.” These AI systems create live 3D digital environments.
The company built two world models called Oasis and Lucy. These tools support gaming, robotics, e-commerce, and simulation work.
World models may become very important in future AI systems. Many experts believe AI agents will need digital spaces where they can test actions, study movement, and solve tasks before they work in the real world.
This area attracts strong investor interest because it connects AI with gaming, automation, and virtual experiences.
The Bigger AI Chip Battle
The Decart funding round arrived during a massive wave of AI infrastructure investment.
Startups and tech giants now fight for control of the AI computing market. Companies want faster chips, lower costs, and better efficiency.
Another startup named Fractile raised $220 million this month for AI inference chips. Cerebras also gained strong investor attention after its public market debut.
At the same time, Amazon and Google continue work on custom AI processors. These firms want alternatives to Nvidia hardware.
This competition may reshape the AI industry over the next few years. Today, Nvidia dominates the market. Tomorrow, AI companies may use many chip systems together.
If that future arrives, Decart could sit at the center of the transition.
Why Investors Feel Excited
Investors see huge opportunity in software that saves time and money for AI labs.
AI development costs continue to rise. Companies spend billions on data centers, chips, and cloud systems. Any tool that reduces friction gains attention very fast.
Decart promises faster product launches and lower migration costs. This value matters because AI firms cannot afford long delays while competitors race ahead.
The startup also fits a larger market trend. AI companies no longer want dependence on one chip supplier. Flexibility now matters almost as much as raw computing power.
That idea explains why major investors rushed into the funding round despite tough market conditions for many startups.
What Comes Next
Decart plans aggressive hiring after the funding round. The company already moved faster on new DOS product releases because customer demand grew quickly.
The startup now stands in a powerful position. It works with large customers, attracts elite investors, and operates inside one of the fastest-growing sectors in technology.
The AI market still changes every month. No one knows which chips will lead five years from now. Yet one fact looks clear today. Companies need simple ways to move AI systems across different hardware platforms.
Decart wants to become the bridge that connects all of them.
Also Read – Nvidia Bets Big on Indian AI Startup Simplismart