2024 Roadmap

Clément Fermaud
Published in
5 min readJan 11, 2024


This year’s roadmap from is not just a list of targets but a blueprint for innovation in DePin (Decentralized Physical Infrastructure). This detailed roadmap aims to dissect the technical aspects and strategic implications of each element in this year’s development. From enhancing GPU support in Virtual Machines to the integration of Oracle and the deployment of Confidential Virtual Machines,’s agenda is set to address critical needs in decentralized computing while paving the way for new industry standards.


Is it a code name for a meme or something bigger? Will you be able to decipher 26 like a true detective?


AWS, Google Cloud, and Azure typically operate on a pay-as-you-go model, where users pay for the resources they consume. This model offers flexibility and scalability, as costs are directly linked to usage.

Our approach is to implement a similar structure in a decentralized environment to provide more transparency and lower costs due to the distributed nature of resources. This system will involve payment streams, in our native token $ALEPH and stablecoins, enabling users to pay for exactly the amount of computing or storage resources they consume, by the millisecond. We are working with a dedicated blockchain and payment stream protocol to offer the best stream solution at a lower cost.

Additional explanation: When a payment is made to rent computing resources, the CRN chosen receives 80% of the payment allocation whether it is in $ALEPH or in stablecoins. The remaining 20% is used as a protocol fee, going to the aleph network to keep incentivizing node operators, pay for network development and other exciting things. If the payment was made in stablecoin then it is automatically converted in $ALEPH to the network.

Example n°1: A user rents computing resources to the CRN “ABC” for a total of $100 paid in $ALEPH. The stream distributes 80$ in $ALEPH to the CRN operator and 20$ in $ALEPH goes to the network.

Example n°2: A user rents computing resources to the CRN “XYZ” for a total of $100 paid in $USDC. The stream distributes 80$ in $USDC to the CRN operator and 20$ in $USDC automatically converted in $ALEPH and goes to the network.


Developing an oracle has meant creating a system that not only fetches data from external sources but also indexes this data, making it easily accessible and usable for decentralized applications. This enhancement has significantly improved the dApp ecosystem on by providing reliable and efficient access to real-world data.

In 2023, we successfully launched our reverse oracle, which allows data to flow from blockchains to the aleph network, essentially functioning as the opposite of a traditional oracle that imports external data into the blockchain.

This year, we are launching our advanced oracle that fetches data from multiple sources, further enhancing reliability and reducing the risk of manipulation. Paired with our decentralized cloud solutions, we plan to increase the security for Real World Assets (RWAs) on-chain.

GPU support for Virtual Machines

Centralized Clouds like AWS, Google Cloud, and Azure offer robust GPU support for virtual machines, catering to high-performance computing tasks like AI, ML model training, inference, and graphics rendering.

Incorporating GPU support in’s decentralized Virtual Machines will make high-performance computing tasks more accessible in the decentralized space. It will democratize access to high-powered computing resources, potentially at a lower cost due to the decentralized nature of resource allocation. GPU support in’s VMs will significantly enhance the network’s capabilities for handling compute-intensive tasks like AI, machine learning, and complex data processing. This is especially pertinent given’s focus on decentralized AI, as seen in projects like

IPFS on Virtual Machines executors

Compute Resource Nodes currently rely on [Core Channel Nodes / other nodes on the network] to obtain the files required to run Virtual Machines. Running their own IPFS node locally, tuned to their requirements, will significantly enhance the reliability, the security and the flexibility of data transfers in the decentralized network.

Confidential Virtual Machines offers a range of computing solutions, including on-demand and persistent virtual machine instances, and soon confidential VMs. A key aspect of these confidential VMs is their ability to run with fully encrypted and isolated memory, storage, and execution. This means that the data processed within these VMs is protected from external access, including from the host system. This level of security is crucial for applications that handle sensitive data or require a high degree of privacy. The decentralized framework not only adds a layer of security through its inherent design but also promotes trustlessness in the system. Users can process data without having to place trust in a central authority, as is the case with traditional cloud services.’s decentralized nature will offer additional benefits in terms of enhanced security and trust minimization.

We also emphasize ease of use in our cloud solutions, offering a user-friendly interface for setting up and managing these VMs, making it accessible for users with varying levels of technical expertise.

We are actively working on confidential computing for meaning that it will support confidential AI inference: no one, not even the node operator will be able to see what is being passed to or generated by a model.

Network Health Status Page and Dashboard

This service typically offers a comprehensive dashboard and status pages for the network, ensuring transparency about system performance and issues. A network health status page for will serve a similar purpose, providing transparency about the decentralized network’s status. This will be critical for trust and reliability, as users will monitor network health in real-time, which is vital for decentralized infrastructure.

New EVM chains supported

Supporting new EVM chains indicates expanding compatibility with different blockchain protocols. This will enhance’s utility in the broader blockchain ecosystem, allowing more dApps and services built on various EVM-compatible chains to integrate and operate on’s platform, especially in the gaming industry.

GeoLocalisation and selection for Storage and Compute Resource Nodes

Centralized Clouds offer region-specific data centers, allowing users to choose where their data is stored and processed for reasons like latency and legal compliance.’s approach to implement geolocation for storage and compute nodes in a decentralized network will be a significant step. It will allow users to select nodes based on geographic location, optimizing for latency and adhering to data sovereignty laws, which is particularly relevant in scenarios where data needs to remain within specific geographical boundaries and hardware features. ecosystem

Thank you and keep in touch.

See you at ETHDenver!

Join our live conversation on our Telegram Community Chat.

🌴 | 🌐 Website | 🗞 Blog | 📄 Papers | 🐦 Twitter | 💬 Telegram |💼 Linkedin | 💻 GitHub | 📒 Dev Docs | 🤖 Reddit