On April 28, 2026, an announcement that seemed "cross-border" brought mining machine manufacturers and token issuers to the same table. Tether announced a strategic partnership with Canaan Inc. (NASDAQ: CAN) and ACME Swisstech to collaboratively develop a new generation of modular, high-density Bitcoin mining infrastructure, using a set of custom computing systems to expand the dedicated computational foundation that Tether is building.
This is not a simple "rebuilding of a mining machine." The publicly disclosed information indicates that the architecture of the new system is intentionally dismantled from the traditional closed, integrated form: computing units, power supply systems, and chassis/enclosures are completely decoupled and exist as independent modules. Canaan provides customized high-density Avalon series hashing board modules for this project, clearly identified as one of the computing cores; ACME Swisstech participates in system integration and engineering, enabling this modular solution to transition from design drafts to tangible engineering products. For Tether, this is a hardware rewrite aimed at "large-scale mining scenarios," directly targeting computational efficiency and cost structure.
In the expression of Tether CTO Paolo Ardoino, this is defined as an active transformation of the traditional "closed, integrated" mining machine paradigm—shifting from all-in-one machines to mining infrastructure centered around modular computing units. In recent years, beyond its existing dollar-pegged token business, Tether has quietly built its Bitcoin mining landscape, and now extends its reach to the hardware level. However, the company has not disclosed the deployment timetable, performance indicators, or funding scale for this system, leaving a series of questions: why would a company known for dollar-pegged tokens want to personally reshape the mining hardware landscape, and what does it truly seek to control in the computational world—just the cost curve, or something deeper?
From Dollar Peg to Computational Mine: Tether's Geographical Expansion
In the narrative of funding in the crypto market, Tether has long stood at the top of the hierarchy—as one of the largest issuers of dollar-pegged tokens globally, it is used to acting as a "reservoir," rather than a downstream water user. However, in recent years, this reservoir has begun to flow downstream: research briefs clearly point out that Tether has continuously expanded its Bitcoin mining business beyond its existing token issuance, and its collaboration with Canaan and ACME Swisstech is not a sudden idea, but an extension of its existing computational layout to a deeper level.
This time, Tether is no longer satisfied with being the "financial partner" of mining sites or the "purchaser" of computational power, but is directly engaging with the hardware structure itself. Official information indicates that this project is defined as strategic cooperation and technological upgrading, aiming to develop new-generation modular, high-density Bitcoin mining infrastructure: expanding Tether's dedicated computational foundation through custom computing systems, focusing on large-scale mining scenarios. The system employs a design that decouples computing units, power supply systems, and chassis; Canaan provides customized high-density Avalon hashing boards as one of the computing cores, while ACME Swisstech participates in system integration and engineering, enabling this modular approach to be realized in practical engineering form.
From a commercial motive perspective, this choice to cross over from financial products to mining infrastructure is hard to be regarded as a mere "trial." The official emphasis is on enhancing computational efficiency and optimizing cost control—pointing to a typical long-term operational logic: when Tether not only holds Bitcoin, nor merely "rents" others' computational power, but instead locks in mining cost curves through a proprietary, customizable computing system, its management of Bitcoin positions shifts from a singular asset-side allocation to a dual layout encompassing both asset and computational sides.
From a cash flow perspective, mining itself provides a revenue stream of computational power that is highly tied to market cycles. For a company originally centered on token issuance fees and asset management income, entering mining infrastructure means adding a layer of "industrial production line" atop its existing financial business—only this production line does not create traditional goods, but rather computational power and Bitcoin. By employing customizable computing systems and a modular architecture, Tether aims to retain control over critical aspects of that production line, attempting to create a closed loop between computational costs, output cadence, and its management of Bitcoin exposure, rather than surrendering full decisions to external mining machine manufacturers and custodian parties.
Deeper changes lie in the rewriting of narrative identity. In the past, Tether was simplified to "dollar-pegged token issuer," with its power boundary understood as financial—responsible for injecting dollar-pegged tokens into the market, after which exchanges, miners, and developers took their respective positions. However, as it begins to invest, design, and promote a modular mining infrastructure that is entirely different from the traditional closed and integrated mining machine form, what Paolo Ardoino describes is no longer just a generational hardware upgrade but a shift in role: from a financial hub to an infrastructure participant mastering the underlying computational power structure.
On this path of migration, Canaan's customized hashing boards and ACME Swisstech's system integration capabilities resemble two tracks: one connects Tether's accumulated experience in Bitcoin mining operations over the past few years, while the other paves the way for its new narrative—"redefining the form and cost structure of computational power"—toward the future. At the end of the track lies not just a massive pool of dollar-pegged tokens, but a forming suite of Tether's own computational foundation.
Tearing Down the Shell: What Does a Modular Mine Look Like?
If we imagine traditional mining machines as a sealed "black box," Tether's current bet is to dismantle this black box into three distinct parts: a purely computational unit responsible for hashing, a power supply system responsible for managing voltage and current, and the chassis previously viewed as an "iron plate sidekick."
The core design concept disclosed by the official sources is to draw a clear dividing line among these three:
The computing unit is no longer welded to the power supply and enclosure but is inserted into the system modularly; the power supply system is designed and optimized separately according to its own logic; and the chassis transitions from a "carry-all" role to more of a replaceable "shell." Their relationship shifts from "a complete machine sold together" to "three components fulfilling their distinct roles with aligned interfaces."
In this dismantled structure, Canaan plays the most prominent and core part: the computing unit.
As a long-term iterative ASIC vendor of the Avalon series, Canaan is not simply transferring complete mining machines to Tether this time, but providing customized high-density Avalon hashing board modules—these hashing boards are positioned squarely at the "heart" of the system. For Tether, what is genuinely being integrated into the new architecture is not complete mining machines, but replaceable, stackable computational power modules; and Canaan's customization involves breaking down its product line originally intended for whole machines into "bare computational power" suited for this modular system.
Once the computational unit is extracted, who will reassemble the power supply, chassis, and these hashing boards back into a system?
The research brief notes that ACME Swisstech is regarded as a critical link on this front—not aiming to compete with Canaan over who understands chips better, but rather standing at the opposite end, responsible for system integration and engineering realization: linking the separated computational modules, power sections, and enclosures into a whole through a complete engineering solution, ensuring this modular framework can operate in real mining environments.
In other words, while Canaan transforms computational power into "pluggable bricks," ACME is expected to be responsible for turning those bricks into walls—making Tether's envisioned "independently replaceable and optimized computation, power supply, and chassis" mining site, more than just a schematic.
It is essential to emphasize that until the announcement of this cooperation on April 28, 2026, Tether had not provided specific computational indicators, power consumption parameters for this system, nor even disclosed where or when it would be deployed. The official emphasis has been on direction:
Using a modular decoupling approach, moving away from the traditional closed, integrated mining machine form, to convert "computational power" from a whole machine into a system that can be composed, replaced, and layered optimized within its own infrastructure. Canaan is responsible for supplying pluggable Avalon computational modules, ACME is responsible for transforming these modules into a system feasible for engineering implementation, and Tether aims to use this approach of tearing down the shell to rewrite its hardware narrative in the Bitcoin mining business.
Who Toppled the Walls of Closed Mining Machines
Over the past decade, Bitcoin mining machines resembled tightly sealed "black boxes": ASIC chips, hashing boards, power supply systems, and enclosures were closely packaged together, with heat dissipation, wiring, and air ducts pre-set by manufacturers. What mining site owners could genuinely do was often just plug in the whole machine and then replace it in bulk when the scrap cycle arrived. An iteration in chip technology meant taking an entire cabinet of old machines offline; if power supply efficiency lagged, few would attempt to repair a fully depreciated machine. The traditional closed, integrated design turned upgrades and maintenance into a blunt capital decision rather than an actionable operational activity.
Paolo Ardoino clearly defines this project as a transformation of such "closed, integrated" forms—not rebuilding a mining machine, but using modular computing units to reconstruct an entire set of mining infrastructure. In the official disclosed plan, computation, power, and chassis are intentionally decoupled: Canaan provides customized high-density Avalon hashing boards as the computing core, inserted into the system designed by ACME; power supply is seen as a module that can be independently upgraded, while the chassis and enclosure's role becomes merely bearing and dissipating heat, rather than being welded into a "exoskeleton." In this architecture, when next-generation Avalon hashing boards are ready, operators can simply replace the computation module while continuing to utilize the existing power supply and chassis; if the power supply technology has new solutions, they can begin iterating from power supplies without having to declare the still-functional hashing boards as obsolete.
In the era of integrated mining machines, the "upgrade" of mining sites was measured in units; after modularization, upgrades were divided into blocks of hashing boards and lines of power supply. For the "large-scale mining scenarios" specifically targeted by the official sources, this granularity of changes directly corresponds to the extent of cost control: a mining site no longer needs to scrap entire racks to keep up with technology nodes but can replace hashing boards as needed, rolling upgrades by region and batch to spread capital expenditures over a longer time frame. The research brief views this decoupled architecture as an important design innovation, believing it has the potential to enhance operational flexibility and cooling efficiency—in a mining site made up of thousands of devices, even just shortening maintenance windows by a few hours or slightly improving heat dissipation could translate into considerable marginal gains over long-term operations.
The real loosening of the walls begins at the point where the power distribution along the industry chain is also redrawn. Closed mining machines bundle computation, power, and chassis into one brand story, with mining site owners purchasing a whole set of "black box capabilities" from a single factory; whereas in Tether's modular narrative, Canaan becomes a pluggable computational supply entity with its high-density Avalon hashing boards, and ACME stands at the system integration nexus, assembling different components into a feasible engineering system. The "binding effect" of complete machine brands is weakened, replaced by the performance, compatibility of individual components, and the capabilities of the system integrator. Mining machines no longer compete solely as complete units, but rather expose the strengths and weaknesses of hashing boards, power supplies, and chassis individually, shifting competition from "which machine is more closed and integrated" to "which module is more efficient and easier to integrate." The walls of closed mining machines are toppled not by a single new machine but by this set of decoupled, reconfigurable infrastructure, piece by piece.
Canaan and ACME: Who Profits in the New Landscape
In the dismantled mining machine structure, Canaan's position has quietly shifted a notch. It is no longer a traditional complete machine manufacturer that sells packaged whole Avalon mining machines to customers; instead, it directly extracts the "heart" to design and supply high-density Avalon series hashing board modules for Tether—this time, it is not machines being sold, but computational power units themselves.
For Canaan, this role transformation is significant. In the entire machine era, it controlled an integrated product from chip to chassis; after modularization, what needs to be delivered to Tether is a hashing board that can be repeatedly plugged in and "reused" with different enclosures and power systems. The product boundary has shrunk, but the bargaining power does not necessarily diminish—as long as Tether successfully runs this architecture, the computational core will become the least replaceable part of the entire system, and this is precisely what Canaan excels at, honed over a long time through the Avalon product line.
ACME Swisstech finds itself pushed to a previously overlooked position: system integrator. The research brief points out that its focus is not on making chips but on reassembling the separated components of Canaan's hashing boards, power supply systems, and enclosures into a functioning whole, ensuring this modular framework is deployable and maintainable in engineering terms. In other words, the engineering details that were previously enclosed under the machine’s chassis lid are now explicitly handed over to ACME.
Once computing, power supply, and chassis are extracted, whoever holds the "assembly manual" controls the narrative. Under this architecture, ACME is tasked with aligning, optimizing, and standardizing different modules—it decides how hashing boards are laid out, how power supply is routed, and in what form the chassis should carry these components. This ability to control the "skeleton" of the entire machine elevates its negotiating position in the mining machine supply chain to a level comparable to that of traditional whole machine manufacturers.
For Canaan, this cooperative model represents an experiment in "carving out its own products": it voluntarily relinquishes complete control over the entire machine form, instead safeguarding the high-value core module of computational power within Tether's newly defined architecture; for ACME, this is an opportunity for advancement, transitioning from a traditional engineering contractor into the headquarters of modular mining machine integration and coordination. Tether has chosen to split value across different tables and then rebind them through contracts and interfaces.
The demonstrative effect is also hidden in this assembly logic. For other mining machine manufacturers, once Tether's modular solution is proven viable in large-scale mining scenarios, it would become challenging for them to continue pretending that the old order of “full machines are everything” remains solid. More manufacturers may be forced to contemplate: should they continue selling closed all-in-one machines, or should they, like Canaan, break out hashing boards, control units, and participate in different projects as components for various mining sites and integrators?
Mining operators will equally be pushed by this cooperation to make choices. In the past, they were faced with complete brand mining machines; now, they have the opportunity to combine hashing boards from different manufacturers, power from different suppliers, and chassis within the same rack, configuring the entire system based on their operational preferences. If the collaboration between Tether, Canaan, and ACME proves this approach effective, operators will have more motivation to demand "replaceable modules" rather than locked-in "complete machines," forcing mining machine manufacturers to open interfaces and dismantle products into negotiable components.
The true beneficiaries may not be a single entity, but players willing to redefine their positions in this round of structural reorganization: Canaan transitions from a full machine manufacturer to a computational module supplier; ACME elevates its bargaining power in the name of system integration; and Tether disperses the structural advantages originally held tightly by whole machine manufacturers into a controlled, reconfigurable modular ecosystem. The next question to be answered is whether other manufacturers and mining sites will choose to observe or follow suit by dismantling and reorganizing their own operations.
The Computational Power War Intensifies: How Far Will Modularity Go?
For Tether, this bet on modular, high-density mining infrastructure is fundamentally not about "buying a few new machines," but rather trying to reshape hardware forms to rewrite its computational cost curve and operational methods. By breaking apart the computing units, power supply systems, and chassis, assigning optimization roles to different entities, and encapsulating computational power into standardized modules with customized high-density Avalon hashing boards, Tether is no longer merely bargaining over existing product price lists but directly intervening in mining machine design, striving to achieve structural advantages in core metrics such as energy density, maintenance efficiency, and downtime.
This also extends Tether's mining business from the traditional path of "purchasing equipment and building machine rooms" to "participating in the design of the equipment itself." Between Canaan's computational modules and ACME's system integration, Tether creates its dedicated computational foundation as a scalable modular base—if the project succeeds, it will own not just a batch of current machines but an entire replicable and iterative hardware paradigm.
However, it must be clear that what the outside world currently sees remains a highly abstract blueprint. The official sources have yet to disclose any specific deployment timetable or mass production plans, nor have they provided expected performance metrics regarding computational power or energy consumption, let alone publicly reveal investment amounts or detailed financial terms. In the absence of these key parameters, any calculations of short-term economic returns or judgments on the alleged "high density" and "high flexibility" can only remain at the level of architectural ideas and directional judgments. Despite the research brief indicating that this modular decoupling is regarded as a significant design innovation in the field of Bitcoin mining with the potential to improve operational flexibility and cooling efficiency, these descriptions are a distance away from being “proven effective,” and the true answers can only be revealed by subsequent operating data.
Moving forward, it feels more like two scripts splitting ahead: if within the next year or two, Tether's pilot projects with Canaan and ACME based on this modular infrastructure successfully land and demonstrate verifiable cost advantages and operational benefits in actual mining scenarios, then today's collaboration may likely spark more modular attempts across the industry—forcing complete machine manufacturers to rewrite product lines and prompting mining sites to plan assets by modules instead of by complete machines, shifting the computational power war from “who's machines are cheaper” to “who's hardware architecture is more flexible.” Conversely, if the cadence of implementation remains vague, and deployment timetables are delayed, with performance and financial metrics remaining blank over the long term, this wave of enthusiasm for modularization may remain locked in the layers of "concepts" and "custom projects," surfacing only in a few deeply-bound collaborations without generating a large-scale industry consensus.
In this moment, Tether's choice has shifted the battlefield from electric prices and procurement costs to the form of mining machines themselves, but how far modularity can reach depends not on declarations but on each hashing board that is physically mounted in racks and each system that is truly operational. For industry participants, what deserves attention is the gradual announcement of deployment plans, performance metrics, and investment scales by officials in the future, and whether these figures will convincingly demonstrate whether modularity truly delivers sufficient results in computational costs and operational efficiency.
Join our community, let’s discuss together, and grow stronger!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
OKX benefit group: https://aicoin.com/link/chat?cid=l61eM4owQ
Binance benefit group: https://aicoin.com/link/chat?cid=ynr7d1P6Z
Disclaimer: This article represents only the personal views of the author and does not represent the position and views of this platform. This article is for information sharing only and does not constitute any investment advice to anyone. Any disputes between users and authors are unrelated to this platform. If the articles or images on the webpage involve infringement, please provide relevant proof of rights and identity documents and send an email to support@aicoin.com. The relevant staff of this platform will conduct an investigation.