MoneyFrontier Summit Insights: When Domestic GPUs Meet the Agent Wave, Four Signals Emerge from the Computing Power Industry
Author: Coolpu News
Compiled by: Heart of Computing Power
The narrative around AI computing power is shifting from "who has the most cards" to "who can truly turn computing power into productivity".
On July 28, the MoneyFrontier 2026 New Infrastructure Intelligent Computing Summit was held at the Hehe Hotel in Hong Kong. Leaders from four companies—Moore Threads, Tencent Cloud, Yanjing Microelectronics, and ChipCool Liquid Cooling—took the stage one after another, revealing the underlying cards of the AI computing power industry chain from four dimensions: the domestic GPU ecosystem, engineering of intelligent agents, the logic of mining transformation, and the reconstruction of data center engineering.
This was not a product launch, but a concentrated presentation of industry consensus: Token consumption is experiencing exponential growth, Agents are evolving from dialogue tools into task execution environments, the low power consumption and energy operation capabilities accumulated in traditional mining are beginning to migrate to AI, and data centers are transitioning from PoW mining rooms to AIDC, which is essentially a fundamental reconstruction of the engineering system.
Here are the key points from the morning speeches at the summit.
1. Moore Threads: Domestic GPUs Enter the "Token Era"
Mr. Zhao Zhenyang, General Manager of the Solutions Department at Moore Threads, delivered a speech titled "Token Era: Everything is Intelligent," sharing insights on the growth of Token demand, the industrialization path of domestic GPUs, and the computing power infrastructure in the era of intelligent agents.
(Figure 1: Guest Zhao Zhenyang from Moore Threads on site)
The speech pointed out that AI is evolving from visual AI and generative AI to Agentic AI and Physical AI. As intelligent agents continuously call upon models, tools, and other agents while executing tasks, the subject of Token consumption is expanding from "humans using AI" to "AI using AI," and the industry's demand for computing power will shift from single-chip performance to system support capabilities for training, inference, simulation, intelligent agent operation, and end applications.
In this context, Moore Threads proposed three frameworks: training factories, token production factories, and intelligent agent factories, corresponding to model training, token production, and intelligent agent operation, respectively. Relevant content indicates that for domestic GPUs to enter real industrial scenarios, it is no longer just a matter of hardware replacement but requires forming a complete closed loop from chips, software stacks, training platforms, inference frameworks, simulation capabilities to application adaptation.
2. Tencent Cloud: Agents Evolve from Dialogue Assistants to Task Execution Environments
Focusing on the technological evolution of AI Agents and their enterprise-level implementation, Mr. Cai Peng, Chief Architect of Tencent Cloud in Yunnan, delivered a keynote speech titled "The Wave of Intelligent Agents: A Comprehensive Reshaping from Technological Eruption to a New Era of Industry."
(Figure 2: Guest Cai Peng from Tencent Cloud speaking on site)
Cai Peng stated that AI Agents are transitioning from large model applications of "you ask, I answer" to a new stage where they can understand goals, break down tasks, formulate plans, call tools, and provide continuous feedback. Once intelligent agents enter industrial scenarios, the core focus is not only on model capabilities but also includes memory mechanisms, knowledge management, tool invocation, skill accumulation, multi-agent collaboration, and secure execution environments.
In terms of enterprise practice, Tencent Cloud showcased solutions such as WorkBuddy, CodeBuddy, ADP intelligent agent development platform, training and inference platform, and AI Sandbox. Relevant content shows that enterprise-level Agents are evolving from personal efficiency tools to task execution, organizational collaboration, and reconstruction of production processes.
3. Dr. Yang Zuoxing: More AI Computing Power Opportunities Occur on the Inference Side
Dr. Yang Zuoxing, founder of Yanjing Microelectronics, brought a keynote speech titled "From Bitcoin Computing Power to AI Computing Power," discussing topics such as cooling routes, energy choices, AI computing power opportunities, and the future of Bitcoin mining.
(Figure 3: Guest Dr. Yang Zuoxing from Yanjing Microelectronics speaking on site)
Dr. Yang believes that for high-density computing power, cooling and energy are no longer just supporting issues but are core variables determining the long-term competitiveness of projects. After comparing air cooling, oil cooling, and water cooling, he judged that water cooling is more suitable for future new construction and renovation projects; on the energy side, natural gas, with its stability and adjustment capabilities, is a noteworthy incremental choice.
Regarding AI computing power, Dr. Yang stated that the training market is becoming concentrated, and the real incremental opportunities are more on the inference side. As AI applications, intelligent agents, and robots become more widespread, inference demand will spread to regional, edge, and industry nodes, while open-source models, distributed deployment, and domestic chips will also create new participation spaces.
Speaking of Bitcoin mining, he candidly stated that the golden age of the industry has passed, but the low-power chips, energy efficiency optimization, and energy operation capabilities accumulated in mining still hold value. In the future, new collaborative models may emerge between mining machines, natural gas, light storage, and AI data centers.
4. ChipCool Liquid Cooling: AIDC Transformation is Primarily an Upgrade of Engineering Systems
Mr. Wang Chong, CEO of ChipCool Liquid Cooling, delivered a keynote speech titled "The Iteration of Computing Power Waves: Opportunities and Real Challenges in the Transformation of the Data Center Industry," focusing on the practical issues of upgrading traditional data centers and blockchain mining rooms to AIDC intelligent computing centers.
(Figure 4: Guest Wang Chong from ChipCool Liquid Cooling speaking on site)
Wang Chong stated that upgrading traditional data centers to AIDC is not simply about replacing GPU servers, but requires an overall reconstruction of power supply and distribution, liquid cooling, networking, fire protection, load-bearing, operation and maintenance platforms, and delivery systems. High-density GPU cabinets are rapidly moving towards tens of kilowatts and even hundreds of kilowatts, which raises higher requirements for power redundancy, liquid cooling control, fire protection standards, and continuous operation and maintenance capabilities.
He pointed out that existing data centers still hold value, and land, factory buildings, power access, and first-side cooling facilities can be reused. However, whether the transformation can be completed depends on whether secondary-side liquid cooling, dual power supply, backup power sources, gas fire protection, network latency control, BCIM operation and maintenance platforms, and standardized quality verification capabilities are in place. For the traditional computing power industry, the AIDC transformation is primarily an upgrade of engineering capabilities.
Four speeches, four perspectives, point to the same conclusion: AI computing power is entering a stage of "system engineering victory."
Moore Threads is working on the full-stack closed loop of domestic GPUs from "being able to run" to "being easy to use"; Tencent is transforming Agents from laboratory toys into enterprise-level production tools; Yang Zuoxing is migrating the energy and engineering experience accumulated in mining over the past decade to inference computing power; and ChipCool Liquid Cooling is upgrading data centers from "powering on" to "zero interruption, verifiable" enterprise-level standards.
This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.
You may also like

Inflation Expectations Rise, Approaching 36% for the Next 12 Months

$7.5 billion raised, $1,306 in daily revenue: The graveyard of ghost blockchains

New Purchasing Regime for Imports by ARCA: Limits, Eligible Products, and Key Points

Elon Musk's xAI Sues Minnesota to Kill the US's First AI Nudification Law

Evernorth updates SEC filing for $1B XRP treasury

Tether Now Has Two Stablecoins; USAT Targets the U.S. Market While USDT Remains Global

Strategy Sells Shares to Pay Dividends? Smells Like a Pyramid Scheme

What It Means to Sleep Covered, Even When It's Hot, According to Psychology

Bitcoin: The Largest Russian Miner Wobbles After Its Founder’s Incarceration

ETF: Capital Flows Back to Bitcoin at the Expense of Ethereum

Kospi: Even a 19-fold profit at Samsung is not enough to calm the panic

ANSES Discounts for Retirees at Supermarkets: How to Access and Which Stores Participate

From Stock Trading to Cryptocurrency Trading: Welcome Back to the Original Family

Central Bank of Russia Introduces Rules for Margin Trading of Cryptocurrencies for Investors

Over 650 trucks stranded in Neuquén due to the closure of the Cristo Redentor Pass and severe weather in the mountains

BTC's Trillion-Dollar Market Value Lies Dormant as 'Awakener' Hashi Testnet Launches Sui

When and at what time will Javier Milei's national address be to announce the BCRA reform

New ARCA Limits: Monthly Billing Allowance for Each Monotributista by Category

The Macroeconomic Logic of the Artificial Intelligence Economy: K-shaped Recovery or a Historical Turning Point?

What Did Stripe See in OpenRouter for a $10 Billion Acquisition?

Bernstein says Core Scientific's AMD partnership could generate $14B over 15 years

The Federal Reserve is Never the Referee

Ripple takes center stage at Wyoming blockchain event

Forget ETF flows, Bitcoin's real threat is a hidden $39,900 liquidation wall

India Orders Removal of Offline Messaging App Bitchat

In-Depth Analysis of FWA: An Interesting Experiment Turning NFTs into "On-Chain Gacha"

Futu not under investigation as Hong Kong SFC freezes HK$125M client assets

Bitget Wallet turns cashback into Bitcoin and stocks

The Trust Trap of Open Protocols: Why Does x402 Need a Centralized Accountability Layer?
