The Polkadot blockchain is a multi-chain blockchain that is designed to bundle multiple chains together so that they can be executed in parallel. The goal is for every block on the Polkadot blockchain to contain multiple blocks from other chains. This will enable the network to contain the transaction information from multiple chains and dramatically increase the overall interoperability between different blockchain networks. Polkadot was created by the Web3 Foundation, which is led by Gavin Wood, one of the founders of Ethereum. Polkadot is their flagship project, and aims to solve the issues in the blockchain ecosystem related to disjointed governance and interoperability.

ArcBlock is creating a blockchain ecosystem for building decentralized applications. Their goal is to provide consumers with the framework and tools to use decentralized apps as easily as current web-based versions. ArcBlock was founded by Robert Mao and Flavien Charlton, who also created Colored Coins, Open Assets Protocol and Openchain.
ArcBlock's Open Chain Access Protocol gives developers an abstract layer for using underlying blockchains. It has a serverless computing architecture that supports various types of applications. ArcBlock is based on the Blocknet application protocol that executes smart contracts and can access external data sources. Blocknet's design and open chain access protocol provide a layer for accessing multiple blockchains.

Seele is developing a blockchain ecosystem that will consist of subnets optimized for different use cases. Seele's Forest Network will have subnets that will operate independently and can interact with each other through cross-chain and cross-domain protocols. Seele has a neural consensus algorithm that was inspired by the inner workings of the human brain. It asynchronously processes and sorts data in a large-scale decentralized environment, using ε-differential agreements to allow all nodes within a network to reach consensus. Compared to existing consensus algorithms, it ensures greater fault-tolerance and is significantly more efficient. Its performance also scales linearly as network size increases, allowing it to be widely adaptable to many different use cases.