Original author: Sleeping in the rain
What exactly is decentralized computing?
In addition to io.net $AKT $AR $TAO, what other opportunities can we participate in?
The following content aims to talk to you about how I understand the decentralized computing track after learning relevant knowledge.
Lets dive in
1. What exactly does decentralized computing do?
That is to say, calculation, what do these protocols need to calculate?
Simply put, computing is the processing of information data to achieve target output results.
The biggest demand side for decentralized computing, or what the market believes is the demand side, is AI training. Of course, this track currently has many problems in data synchronization, network optimization, and data privacy and security.
At present, the biggest solutions on the market should be io.net $AKT and $RNDR. However, as Greythorn Asset Management mentioned, the complexity of creating and managing decentralized clusters on a large scale involves significant technical challenges, and they still have a long way to go (it is for io.net, but it applies to most.
Reference link: https://0x greythorn.medium.com/io-nets-revolutionary-gpu-cloud-f18c06b944e4
2. Refine the application direction
Let’s take a look at the specific business of the decentralized computing project mentioned earlier ⬇️
io.net focuses on GPU computing and has cooperated with $RNDR, with application directions including machine learning and artificial intelligence computing.
$TAO acts as a computing power intermediary to match the needs of artificial intelligence training.
$AKT is more scalable than $RNDR. It supports GPU, CPU and storage computing, so its customer base is also richer.
AO launched by $AR is a super parallel computer that is modularly combined with Arweave. AO is responsible for communication and parallel computing, while Arweave is responsible for storage and verification. In addition, AO supports modular combination.
The application direction of decentralized computing is strongly linked to the AI field, providing services to the AI field in the form of computing power. In essence, we can break down Crypto+AI into: 1) What can Crypto do for the AI industry? 2) How can the AI industry empower Crypto? I have mentioned this in previous articles. The way the AI industry empowers Crypto is through AI agents, such as $PRIME $OLAS. The basic idea of Crypto being able to serve the AI industry is to provide computing power.
This is also the reason why computing power targets are being hyped and new computing power protocols are emerging.
In addition to computing power and applications, Crypto can also do something at the data and algorithm levels.
At present, their main growth bottleneck lies in the acceptance of their cooperation form by Web2 customers. In this regard, $AKT is doing relatively well.
3. Two Gems related to computing and data
Next, I will share with you a few Gems (I have positions in all of them and have a vested interest. Please do your own research before buying and don’t take over from me).
FluenceDAO @fluence_project
Official introduction: Fluence is a Web3 native computing platform for developing and hosting applications, interfaces, and backends on a permissionless peer-to-peer network. Fluence can read data from any public data source (IPFS, Filecoin, Arweave, Ceramic, Ethereum, Solana, Flow, etc.), calculate it, and store the newly calculated data back in any of these repositories.
Background: FluenceDAO is an AI+DePin project that has reached a cooperation with Filecoin. Solana Co-founders have also paid attention to this project. The project is led by Multicoin, with participation from 1kx and Signum Capital, and the financing amount is US$11 million. Fluence has created a network to provide users with a decentralized cloud-free computing platform + market, which is managed by Fluence DAO and $FLT.
Currently, the price of $FLT is 0.6 USD, MC is 29.9 M, and FDV is 599 M.
Read more: https://twitter.com/ahboyash/status/1770333758522323192
AIOZ @AIOZNetwork
Official introduction: AIOZ Network is a comprehensive infrastructure solution for Web3 storage, decentralized AI computing, live streaming, and video on demand (VOD). AIOZ Networks dCDN platform changes file storage and distribution in Web 3.0 Dapps, providing an affordable solution for file storage and media streaming. AIOZ Networks blockchain combines the robustness of Cosmos and the compatibility of the Ethereum Virtual Machine (EVM) (high compatibility and low cost).
Background: Previously, AIOZ focused on becoming the main DePin infrastructure platform for storage and streaming media, but now AIOZ is on the road to AI. Just like io.net and FluenceDAO, it is doing AI+DePin infrastructure. A few years ago, AIOZ joined the NVIDIA Inception program.
A special design of AIOZ is dCDN (distributed content delivery network), where edge nodes of the network are responsible for running the network, and nodes that run the network will be able to receive $AIOZ token rewards. One feature of dCDN is that it allows the network to expand infinitely. In other words, as demand grows, the number of edge nodes needs to grow to meet market demand (currently there are 80,000 nodes worldwide).
So, how does AIOZ combine with AI?
AIOZ W 3A I is an AI computing infrastructure that can help customers perform distributed AI computing and ensure data privacy. Customers can access more AI models through the AI-as-a-service service provided by AIOZ.
Interestingly, while reading the materials, I also noticed a concept that was mentioned many times: AI reasoning.
In the field of AI, inference is the process of using a trained machine learning model* to draw conclusions from new data. AI models that are capable of inference can do so without examples of the desired results. Simply put, AI training is the first stage of an AI model, and AI inference is the application of an AI model. In fact, AI inference is the process used to test the capabilities of an AI model.
AIOZs W3A I Marketplace allows nodes to store user data in a decentralized manner and execute AI tasks directly on user devices, which will make AI reasoning more cost-effective and more private.
To put it simply, AIOZ is providing services for AI through edge computing.
Read more: https://route2fi.substack.com/p/aioz-network-at-the-forefront-of
Currently, the price of $AIOZ is 0.8 USD, MC 878 M, and FDV 878 M.
Finally, lets talk about what I think is the trend of Crypto AI: An important trend in the future of Crypto AI is the refinement of the subdivision track, and the granularity will be higher. While competing, more modular cooperation will also come.
