IREN Ltd (IREN) Stock Analysis and Metrics

Iren

NASDAQ: IREN

$38.95 ▲ +0.36 (+0.93%)

At close: Jul 15, 2026 · 11:38 AM UTC

Financial Ratios

Market Cap14.46 Bn

P/E91.47

P/S19.10

Div. Yield0.00

ROIC (Qtr)0.00

Revenue Growth (1y) (Qtr)-0.02

Add ratio to table…

About

IREN Limited owns and operates next generation data centers powered entirely by renewable energy, either from direct clean sources or through the purchase of renewable energy certificates. The company’s facilities are located in Texas and British Columbia and support a combined capacity of approximately 810 megawatts as of June 30 2025. These sites host bitcoin mining operations that rely on application specific integrated circuits and also provide high performance computing and artificial intelligence services using graphics processing units. By owning the land, the electrical infrastructure and the buildings, IREN maintains direct control over its power supply and hardware, which it says leads to more stable cash flows compared with operators that depend on third party hosting or short term leases.

IREN Limited generates revenue primarily from two activities. First, its bitcoin mining operation earns block rewards and transaction fees by contributing hashrate to global mining pools, and the earned bitcoin is exchanged each day for U. S. or Canadian dollars on platforms such as Kraken and Coinbase, with the company typically liquidating all mined bitcoin within the same day to avoid holding the cryptocurrency on its balance sheet. Second, its high performance computing and artificial intelligence business provides cloud based access to graphics processing units, charging customers for use of the computing power in artificial intelligence training and inference workloads, and it offers contract terms that range from on demand to multi year agreements.

IREN Limited operates with vertical integration as an owner and operator of data centers that source all electricity from renewable resources, a combination that sets it apart from many competitors who lease space or depend on third party power providers. In the bitcoin mining sector it faces competition from firms such as Bitfarms Ltd., Marathon Digital Holdings, Riot Platforms, CleanSpark Inc., Core Scientific Inc., Hut 8 Corp. and TeraWulf Inc., while in the high performance computing and artificial intelligence market it contends with cloud providers including Amazon Web Services, CoreWeave, Lambda Inc., Microsoft Azure, Google Cloud and specialized players such as Crusoe Cloud. The company’s competitive advantages stem from its direct ownership of land, electrical infrastructure and mining hardware, its long term power contracts that lock in low cost renewable energy, and its flexibility to allocate capacity between mining and GPU based services according to market conditions. Additionally, IREN’s focus on geographic diversification across the United States and Canada helps reduce exposure to any single regulatory regime or weather event.

IREN Limited’s bitcoin mining revenue comes from the sale of bitcoin obtained through mining pools to cryptocurrency exchanges such as Kraken and Coinbase, while its high performance computing and artificial intelligence revenue is derived from agreements with third party customers who require access to graphics processing units for artificial intelligence training and inference. Although the filing does not disclose specific customer names, the service serves a broad range of users including technology companies, research institutions and enterprise clients that need scalable GPU compute for workloads such as model training, hyperparameter tuning and real time inference. The business reports a mix of contract durations, with some customers opting for short term arrangements and others committing to longer periods of up to three years, which helps the company manage utilization of its infrastructure.

Investment Thesis

▲ Bull case

IREN Limited's NVIDIA partnership represents a structural shift in the AI infrastructure market rather than a temporary arrangement, as the $2.1 billion NVIDIA investment vests only upon deployment of 600,000 GPUs across IREN's campuses, directly aligning NVIDIA's capital with execution milestones and eliminating the risk of passive financial investment. This creates a powerful incentive for NVIDIA to actively support IREN's buildout, leveraging their technological expertise and market influence to accelerate customer acquisition and de-risk deployment timelines, particularly for the $3.4 billion AI cloud contract supporting NVIDIA's internal workloads, which serves as a foundational anchor for broader ecosystem growth. The partnership's design ensures that as IREN scales its platform—targeting 5 gigawatts of secured power globally—NVIDIA's vested interest grows commensurately, transforming what could be a supplier relationship into a co-dependent growth engine where IREN's execution capabilities are amplified by NVIDIA's strategic commitment, thereby reducing execution risk and enhancing long-term revenue visibility beyond current market expectations.
The Mirantis acquisition adds a critical, underappreciated layer to IREN's go-to-market strategy by providing immediate access to 650 engineers with over a decade of experience managing cloud infrastructure for 1,500 enterprise customers globally, directly addressing the unspoken challenge of post-deployment support in complex AI environments—a gap that pure hardware builders often overlook. While management highlighted Mirantis' Kubernetes and AI infrastructure orchestration capabilities, they did not emphasize how this team enables IREN to move beyond bare metal GPU delivery into full-stack AI cloud services, including provisioning, monitoring, and multi-environment support (bare metal, VMs, Kubernetes), which is essential for retaining enterprise and sovereign clients who require higher touchpoints and customization. This capability allows IREN to capture higher-margin, longer-term contracts with AI-native and enterprise segments that hyperscalers alone cannot satisfy, diversifying the customer base and reducing reliance on volatile hyperscape demand, while simultaneously strengthening the NVIDIA partnership through their founding ISV status in the NVIDIA AI Cloud Ready initiative, creating a synergistic flywheel where Mirantis' enterprise reach fuels NVIDIA-aligned deployments and vice versa.
IREN's strategy of retrofitting existing air-cooled data centers—such as the 180 megawatts in British Columbia and Childress, and the 250 megawatts planned for Childress in 2027—represents a capital-efficient, high-velocity path to ARR conversion that the market is underestimating, as it leverages already-permitted sites with existing grid connections, operational teams, and site control, bypassing years of permitting and construction delays inherent in greenfield builds. This approach allows IREN to monetize capacity faster—air-cooled deployments can come online in months versus 18-24 months for liquid-cooled facilities—directly addressing the market's paramount metric of time to compute, which management repeatedly cited as the key driver of revenue conversion. With $700 million of ARR already tied to the 60-megawatt Blackwell deployment at Childress and the company targeting $3.7 billion ARR by end-2026 across 150,000 GPUs, the ability to rapidly convert underutilized infrastructure into contracted AI cloud capacity creates a compounding advantage: each retrofitted site not only generates immediate revenue but also builds institutional knowledge that accelerates future phases, making the 2027 expansion to 1.21 gigawatts more achievable and less capital-intensive than peers attempting similar scale from scratch.

IREN Limited's NVIDIA partnership represents a structural shift in the AI infrastructure market rather than a temporary arrangement, as the $2.1 billion NVIDIA investment vests only upon deployment of 600,000 GPUs across IREN's campuses, directly aligning NVIDIA's capital with execution milestones and eliminating the risk of passive financial investment. This creates a powerful incentive for NVIDIA to actively support IREN's buildout, leveraging their technological expertise and market influence to accelerate customer acquisition and de-risk deployment timelines, particularly for the $3.4 billion AI cloud contract supporting NVIDIA's internal workloads, which serves as a foundational anchor for broader ecosystem growth. The partnership's design ensures that as IREN scales its platform—targeting 5 gigawatts of secured power globally—NVIDIA's vested interest grows commensurately, transforming what could be a supplier relationship into a co-dependent growth engine where IREN's execution capabilities are amplified by NVIDIA's strategic commitment, thereby reducing execution risk and enhancing long-term revenue visibility beyond current market expectations.
The Mirantis acquisition adds a critical, underappreciated layer to IREN's go-to-market strategy by providing immediate access to 650 engineers with over a decade of experience managing cloud infrastructure for 1,500 enterprise customers globally, directly addressing the unspoken challenge of post-deployment support in complex AI environments—a gap that pure hardware builders often overlook. While management highlighted Mirantis' Kubernetes and AI infrastructure orchestration capabilities, they did not emphasize how this team enables IREN to move beyond bare metal GPU delivery into full-stack AI cloud services, including provisioning, monitoring, and multi-environment support (bare metal, VMs, Kubernetes), which is essential for retaining enterprise and sovereign clients who require higher touchpoints and customization. This capability allows IREN to capture higher-margin, longer-term contracts with AI-native and enterprise segments that hyperscalers alone cannot satisfy, diversifying the customer base and reducing reliance on volatile hyperscape demand, while simultaneously strengthening the NVIDIA partnership through their founding ISV status in the NVIDIA AI Cloud Ready initiative, creating a synergistic flywheel where Mirantis' enterprise reach fuels NVIDIA-aligned deployments and vice versa.
IREN's strategy of retrofitting existing air-cooled data centers—such as the 180 megawatts in British Columbia and Childress, and the 250 megawatts planned for Childress in 2027—represents a capital-efficient, high-velocity path to ARR conversion that the market is underestimating, as it leverages already-permitted sites with existing grid connections, operational teams, and site control, bypassing years of permitting and construction delays inherent in greenfield builds. This approach allows IREN to monetize capacity faster—air-cooled deployments can come online in months versus 18-24 months for liquid-cooled facilities—directly addressing the market's paramount metric of time to compute, which management repeatedly cited as the key driver of revenue conversion. With $700 million of ARR already tied to the 60-megawatt Blackwell deployment at Childress and the company targeting $3.7 billion ARR by end-2026 across 150,000 GPUs, the ability to rapidly convert underutilized infrastructure into contracted AI cloud capacity creates a compounding advantage: each retrofitted site not only generates immediate revenue but also builds institutional knowledge that accelerates future phases, making the 2027 expansion to 1.21 gigawatts more achievable and less capital-intensive than peers attempting similar scale from scratch.

▼ Bear case

IREN Limited's aggressive capital allocation strategy, highlighted by the recent announcement of a $2 billion convertible debt offering due in 2033, introduces significant refinancing risk and potential dilution that the market is beginning to price in, as the company's transition from Bitcoin mining to AI Cloud remains incomplete and heavily reliant on external financing despite claiming $2.6 billion in cash at April 30, 2026. While management frames financing as progressive and tied to execution—citing the Microsoft contract's 95% prepayment-funded GPU CapEx—the scale of the NVIDIA partnership (targeting 5 gigawatts and 600,000 GPUs) implies CapEx requirements that far exceed current cash flows and operating revenue, with the $3.1 billion contracted ARR representing only a fraction of the capital needed to build and sustain such infrastructure. The reliance on GPU financing and prepayments creates vulnerability to shifts in customer creditworthiness or market sentiment; if demand for AI cloud softens or hyperscalers delay deployments, IREN could face stranded capital in GPUs or data centers, exacerbating losses from ongoing Bitcoin mining decommissioning impairments (which totaled $140.4 million in the March quarter) and pressuring adjusted EBITDA, which declined to $59.5 million from $75.3 million quarter-over-quarter despite AI cloud revenue growth.
The company's expansion into Europe via the Nostrum acquisition in Spain, while framed as a strategic move into a supportive regulatory environment with abundant renewables, overlooks critical execution risks tied to grid interconnection timelines and local permitting complexities that management did not adequately address when questioned about Spain's 490 megawatts of secured power—Kent Draper noted that final design and permitting are "already well advanced" but offered no concrete timelines or milestones, suggesting potential delays that could erode the first-to-compute advantage IREN emphasizes as critical to revenue conversion. Unlike Texas or British Columbia, where IREN has established operational expertise and streamlined processes, Spain represents a new jurisdictional hurdle with differing grid operator procedures (via REE instead of ERCOT or AESO), potential labor regulations, and less proven experience in rapid data center buildout, meaning the assumed speed-to-market advantage may not transfer seamlessly, especially as the company seeks to replicate its North American modular design in a region with historically slower infrastructure development cycles.
IREN's heavy reliance on hyperscaler contracts—exemplified by the $3.4 billion NVIDIA AI cloud deal and Microsoft collaboration—creates concentration risk that the market may be underappreciating, as these agreements, while large in scale, often come with lower margins due to the wholesale nature of hyperscale purchasing and limited pricing power for infrastructure providers, a dynamic Kent Draper implicitly acknowledged when noting that servicing end customers yields better margins but that the company is not "going 100% hyperscale." The company's stated goal of diversifying into enterprise, AI-native, and sovereign workloads remains aspirational, with no concrete contracts disclosed beyond the NVIDIA and Microsoft deals, and the Mirantis acquisition—while intended to enable this shift—has not yet demonstrated traction in securing non-hyperscale revenue, leaving the bulk of near-term ARR growth tied to a small number of counterparties whose capital expenditure plans are notoriously cyclical and subject to sudden pauses during economic downturns or AI hype cycles, potentially leaving IREN with overbuilt capacity and underutilized assets if hyperscaler demand fluctuates.

IREN Limited's aggressive capital allocation strategy, highlighted by the recent announcement of a $2 billion convertible debt offering due in 2033, introduces significant refinancing risk and potential dilution that the market is beginning to price in, as the company's transition from Bitcoin mining to AI Cloud remains incomplete and heavily reliant on external financing despite claiming $2.6 billion in cash at April 30, 2026. While management frames financing as progressive and tied to execution—citing the Microsoft contract's 95% prepayment-funded GPU CapEx—the scale of the NVIDIA partnership (targeting 5 gigawatts and 600,000 GPUs) implies CapEx requirements that far exceed current cash flows and operating revenue, with the $3.1 billion contracted ARR representing only a fraction of the capital needed to build and sustain such infrastructure. The reliance on GPU financing and prepayments creates vulnerability to shifts in customer creditworthiness or market sentiment; if demand for AI cloud softens or hyperscalers delay deployments, IREN could face stranded capital in GPUs or data centers, exacerbating losses from ongoing Bitcoin mining decommissioning impairments (which totaled $140.4 million in the March quarter) and pressuring adjusted EBITDA, which declined to $59.5 million from $75.3 million quarter-over-quarter despite AI cloud revenue growth.
The company's expansion into Europe via the Nostrum acquisition in Spain, while framed as a strategic move into a supportive regulatory environment with abundant renewables, overlooks critical execution risks tied to grid interconnection timelines and local permitting complexities that management did not adequately address when questioned about Spain's 490 megawatts of secured power—Kent Draper noted that final design and permitting are "already well advanced" but offered no concrete timelines or milestones, suggesting potential delays that could erode the first-to-compute advantage IREN emphasizes as critical to revenue conversion. Unlike Texas or British Columbia, where IREN has established operational expertise and streamlined processes, Spain represents a new jurisdictional hurdle with differing grid operator procedures (via REE instead of ERCOT or AESO), potential labor regulations, and less proven experience in rapid data center buildout, meaning the assumed speed-to-market advantage may not transfer seamlessly, especially as the company seeks to replicate its North American modular design in a region with historically slower infrastructure development cycles.
IREN's heavy reliance on hyperscaler contracts—exemplified by the $3.4 billion NVIDIA AI cloud deal and Microsoft collaboration—creates concentration risk that the market may be underappreciating, as these agreements, while large in scale, often come with lower margins due to the wholesale nature of hyperscale purchasing and limited pricing power for infrastructure providers, a dynamic Kent Draper implicitly acknowledged when noting that servicing end customers yields better margins but that the company is not "going 100% hyperscale." The company's stated goal of diversifying into enterprise, AI-native, and sovereign workloads remains aspirational, with no concrete contracts disclosed beyond the NVIDIA and Microsoft deals, and the Mirantis acquisition—while intended to enable this shift—has not yet demonstrated traction in securing non-hyperscale revenue, leaving the bulk of near-term ARR growth tied to a small number of counterparties whose capital expenditure plans are notoriously cyclical and subject to sudden pauses during economic downturns or AI hype cycles, potentially leaving IREN with overbuilt capacity and underutilized assets if hyperscaler demand fluctuates.

Geographical Breakdown of Revenue (2025)

Peer Comparison

Companies in the Capital Markets

S.No.	Ticker	Company	Market Cap	P/E	P/S	Total Debt (Qtr)
1	MS	Morgan Stanley	330.70 Bn	0.00 Bn	4.50	119.83 Bn
2	GS	Goldman Sachs Group Inc	309.79 Bn	0.00 Bn	5.12	259.45 Bn
3	SCHW	Schwab Charles Corp	167.21 Bn	0.00 Bn	6.74	-
4	FUTU	Futu Holdings Ltd	111.36 Bn	85.66 Bn	82.13	0.01 Bn
5	HOOD	Robinhood Markets, Inc.	97.69 Bn	0.00 Bn	21.18	-
6	LPLA	LPL Financial Holdings Inc.	23.49 Bn	0.00 Bn	1.29	-
7	TW	Tradeweb Markets Inc.	21.59 Bn	0.00 Bn	9.99	-
8	CRCL	Circle Internet Group, Inc.	15.14 Bn	0.00 Bn	6.85	-

Research Tools

Developers

About

Investment Thesis

Product and Service Breakdown of Revenue (2025)

Geographical Breakdown of Revenue (2025)

Peer Comparison

Research Tools

Developers

About

Investment Thesis

Product and Service Breakdown of Revenue (2025)

Geographical Breakdown of Revenue (2025)

Peer Comparison

Create an account

Sign In

Reset Password