Optimum, a memory layer that improves the decentralized performance of any blockchain, has raised a $11 million seed round and has invited creators from institutions such as Harvard and MIT to jump from the academia world to commercial crypto arenas.
The seed round was led by 1KX with participation from robot ventures, Finality Capital, Spartan, CMT Digital, SNZ, Triton Capital, Big Brain, CMS, Longhash, NGC, Animoka, GSR, Karadan, Reforge and more.
According to a press release, the best ones will build what is called the missing memory layer of blockchain, creating a faster, cheaper, truly decentralized way of storing, accessing, propagating data.
The core of Optimum’s innovation is the distributed coding method for distributed systems known as Random Linear Network Coding (RLNC), developed by MIT professor Muriel Médard, speaking at Consensus Stronto 2025.
“When you think of Web3 as a decentralized world computer, people do a great job in the computing part. Let’s look at the operating system, for example,” Medard said in an interview. “But anyone who puts a computer together knows that it also needs a bus that is a data propagation, and that, in contrast to more static memory, such as disks and clouds, we need a memory called random access memory.”
Without a scalable memory layer, blockchains face systemic inefficiencies, including outdated gossip networks that propagate data redundantly, crowded meme pools that cause unpredictable delays, and bloated nodes that make searches costly and complicated.
Optimum’s memory infrastructure uses Médard’s RLNC coding scheme to tackle inefficient data propagation, redundant storage, and slow access.
Optimum is currently live on a private testnet, inviting L1, L2S, validators and node operators to operate its distributed memory layer.
