More than 2,300 clients paid to rent GPUs, and the Bittensor subnet achieved an annualized scale of 34.6 million dollars.

CN
3 hours ago

AI encryption projects often face a problem: the amount of attention the network receives and the number of transactions generated by tokens do not directly prove that they are truly being used.

The GPU leasing subnet Lium under Bittensor recently released a set of data that is closer to actual business metrics.

According to Lium's operational data covering October 10 to November 9, the platform generated approximately $2.7 million in GPU leasing expenses within 31 days, had 2,394 paying customers, and completed around 29,300 leases. Compared to the previous month, leasing expenses grew by about 87%, and the number of paying customers increased by approximately 72%.

Based on the expenditure rate over the past week, Lium's GPU leasing annualized operating scale reached approximately $34.6 million.

This set of data provides a more concrete observation metric for the AI computing business within the Bittensor ecosystem: in addition to token incentives, there are already customers willing to pay for GPU computing resources. More than 2300 customers paid to rent GPUs, Bittensor subnet achieved an annualized scale of $34.6 million_aicoin_img1

Where does the $2.7 million leasing expense come from?

Lium is a GPU leasing service running on Bittensor and belongs to Subnet 51 in the network. It aggregates computing power provided by different GPU suppliers, allowing users to rent corresponding hardware resources according to their needs.

For users who need to run AI models, perform inference, or handle computing tasks, GPU resources are a real infrastructure requirement. Lium offers a connection and leasing entry for computing resources.

Among the data released this time, what is most noteworthy is not just the $2.7 million expenditure, but also the number of customers and the number of leases.

Within 31 days, the platform had 2,394 paying customers and completed about 29,300 leases. Meanwhile, leasing expenses grew from approximately $1.45 million in the previous month to $2.7 million, an increase of about 87%; the number of paying customers increased by about 72%, and the number of leases grew by about 62%.

This indicates that the platform's growth is reflected not only in single financial metrics, but customer numbers and actual leasing activities have also increased concurrently.

However, this data still only represents operational performance within a specific time window and cannot directly deduce long-term customer retention rates, nor imply that all GPU resources maintain high utilization rates.

How does Bittensor turn computing power into a market?

The Bittensor behind Lium is not a cloud computing platform operated by a single traditional company.

Bittensor uses a subnet mechanism to split different types of AI and digital services into independent markets. Each subnet operates around specific tasks, participants provide services, and the network evaluates and rewards participants based on rules.

In the GPU leasing market where Lium is located, computing power providers are responsible for providing GPU resources, and users pay fees to rent computing power; validators inspect and evaluate the hardware, GPUs, network connections, and operational capabilities of nodes according to network rules.

This model aims to organize decentralized computing power supply, allowing different suppliers to participate in the same market without having to rely entirely on a single cloud service provider.

Compared to traditional cloud computing platforms, the core challenge of decentralized computing networks is how to ensure resource quality, service stability, and supply-demand matching. Users need not only to connect to GPUs, but also require that the computing power operates as needed, and have verifiable service standards in case of issues.

The number of paying customers and leases at Lium at least provide an observable demand-side metric.

$34.6 million annualized scale does not equal annual revenue

The easiest figure to misinterpret from this data is $34.6 million.

This is not the annual revenue that Lium has achieved, but rather an annualized projection based on the GPU leasing expenditure rate over the last 7 days. In other words, if the same expenditure rate is maintained over the next year, it could theoretically reach this scale.

The actual situation might be affected by factors such as GPU prices, customer demand, hardware supply, market competition, and utilization rates of computing power, so the annualized operational scale cannot be directly treated as realized revenue.

Additionally, GPU leasing expenditures do not equate to platform profits. Computing power providers need to bear costs for hardware procurement, hosting, electricity, and maintenance, while the platform itself incurs operational and service costs. How the rental fees paid by customers are distributed among suppliers, platforms, and other participants will also impact the final economic benefits.

The value of TAO still requires a layer of verification

The growth of Lium provides a noteworthy case for Bittensor, but it should not be simply understood that an increase in GPU leasing expenditures entails an inevitable rise in the value of the TAO token.

Bittensor distributes network rewards through a subnet incentive mechanism, allowing subnet participants to receive corresponding incentives according to rules. There are still various transmission links between real business demand, subnet revenue, token incentives, and market prices.

For TAO, what truly needs observation is: whether more subnets can continuously generate external paid demand, whether users will reuse the service, whether business income can cover the costs of computing power supply and operations, and whether network incentives can form a sustainable economic cycle.

In the past, AI encryption projects have often been questioned regarding their reliance on token subsidies to maintain superficial activity. The paying customers and GPU leasing expenses released by Lium this time provide a different data perspective compared to token prices and network incentives.

However, growth in one subnet is still not enough to prove that the entire decentralized AI network has found a mature business model. Moving forward, customer retention, genuine profits, and whether other subnets can replicate similar demand will be the key to determining whether Bittensor can achieve sustainable development.

If you are also paying attention to the crypto market and want to participate in trading or asset allocation, you can register through the following exclusive channels to enjoy discounts:

Binance: Register through the exclusive link and fill in the invitation code aicoin668 to enjoy a 10% rebate.

Registration link: https://jump.do/zh-Hans/xlink-proxy?id=3

Benefit group: https://www.aicoin.com/link/chat?cid=gmLgwvKD1

OKX: Register through the AiCoin exclusive link to enjoy a permanent 20% rebate.

Registration link: https://jump.do/zh-Hans/xlink-proxy?id=2

Benefit group: https://aicoin.com/link/chat?cid=l61eM4owQ

Bitget: Register through the AiCoin exclusive channel to enjoy a 10% fee rebate.

Registration link: https://jump.do/zh-Hans/xlink-proxy?id=6

Benefit group: https://aicoin.com/link/chat?cid=N63yYYKQE

You are also welcome to join AiCoin-related communities for more information on new coins and earning strategies from exchanges:

Official Telegram community: https://t.me/aicoin_support

AiCoin Chinese Twitter: https://x.com/AiCoinzh

The content of this article reflects the personal views of the author and does not represent the position of this platform. The opinions, conclusions, and suggestions in the article are for reference only and do not constitute any investment advice related to this platform. The market has risks, and investments should be made cautiously.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到[email protected],本平台相关工作人员将会进行核查。

Share To
APP

X

Telegram

Facebook

Reddit

CopyLink