IREN Bull

$IREN

[DNV999 @DNV99999]

Unless Antrhopic wants to wait years, they have to turn to IREN, because the others have already rented almost everything, $CIFR only has 70MW to rent, $NBIS is late on a contract, $CRWV has no MW, $APLD is all rented, the only one with SO MANY MW to rent NOW is $IREN

[Serenity @aleabitoreddit]

New Anthropic news looks like a potential tailwind for the Neocloud colo sector.

Such as $WULF, $CIFR, $WYFI, $HUT and others (not named yet).

As Anthropic is pursuing its first DC leases.

"The AI company has signed more than a dozen letters of intent with U.S. developers" per The Information.

[Mirantis @MirantisIT]

Imagine running production AI at scale, with full control, zero vendor lock-in, and built-in governance from day one.

That's what a Private Enterprise AI Factory makes possible.

In this webinar, Martin Stadler, Field CTO for AI Infrastructure, shows platform teams exactly how to:
• Standardize operations across VM, container & GPU workloads
• Build a composable AI platform that adapts as tech evolves
• Bridge existing enterprise infra with production-ready AI ops

Save your spot now: https://t.co/Grr4iF5ztI

#PrivateAI #EnterpriseAI #AIInfrastructure #AIFactory

[Jim Liu @jiahanjimliu]

If Anthropic is is pursuing it’s first datacenter leases, you can bet that it will bring its own software any Neocloud deals.

Owning the full stack and not interoperating between eight different cloud platforms is how you reduce bugs and increase iteration speed. For Anthropic this means agentic environment down to the bare metal.

[zerohedge @zerohedge]

*CRUSOE PUSHED ASIDE AT WYOMING AI PROJECT AFTER GOOGLE CONCERNS

First LLM price cuts, now data center cancelations

[Frans Bakker @FransBakker9812]

$IREN is currently deploying a similar design in Horizon 😎

"flexible rack densities" meant nothing more than running GB300s in pre-determined VR200 DSX "pods" 👀

My intel suggests the DSX setup is also being mapped out in foundations at SW1 🏗

Look and decide for yourself 🔍

[Jim Liu @jiahanjimliu]

We have objective, facts based reporting on NBIS’s Lappeenranta DC from @TheUglyBuckling.

[Mirantis @MirantisIT]

Nobody knows what AI hardware looks like in 5 years. And if your infrastructure strategy requires ripping and replacing every time one layer shifts, you will be in a constant state of disruption.

The answer is not picking the right vendor. It is building for composability so your organization can evolve without the cost of chaos.

Watch Mirantis SVP of Business Development Richard Borenstein break down why composability is the architectural answer to a landscape in permanent beta, and what that means for enterprises building AI infrastructure today. https://t.co/le1Qnob7zD

#NeoCloud #AIInfrastructure #AICloud #SovereignAI #k0rdentAI

[The Tech Investor @TheTechInvest]

$IREN could 4x by next year for these reasons:

1. Sweetwater 1 (1.4GW) was energized on time.

2. The company doubled its power capacity (2.9GW to 5.8GW; now the largest neocloud in the world surpassing $CRWV and $NBIS)

3. Announced building Oklahoma (1.6GW) data center.

4. Announced its first data center in Australia (800MW).

5. Raised its ARR from $3.4B to $4.4B.

6. Acquired Mirantis, a strategic partner of $NVDA.

7. Announced its expansion in Europe.

8. Announced its strategic partnership with $NVDA to accelerate deployment of up to 5GW of AI Infrastructure.

9. Signed a $3.4B deal with $NVDA.

10. $NVDA to invest up to $2.1B in $IREN (still an option conditional upon deployment of 600K GPUs.)

11. Sweetwater (2GW) selected as a flagship deployment for $NVDA DSX AI factory architecture.

12. Disclosed having a global pipeline.

[Shanu Mathew @ShanuMathew93]

Oof looks like Crusoe got the boot? "The company is continuing to advance the project and is working directly with the potential large-load customer. While Crusoe is no longer the development partner for this project, the project continues to progress as planned and is tracking for service to begin in early 2028."

[Frans Bakker @FransBakker9812]

Where I appreciate the positive view on this, I still think that Crusoe will have to answer for this apparent fumble.

Nothing that I took from any of the reports, says that there is a "lack of demand for AI".

The nuance from Shanu is political correct, but it is most likely a issue with the combination of $BE cells, and the way @CrusoeAI can adopt this into their data centers.

From what I heard:
1. There was a massive error in procuring gas turbines for Abilene, this lies with Lancium/Crusoe, so at a first major project coming in with this, is a major red flag to me.
2. Their Childress project is already locally mentioning Oracle as the customer. Subcontractors working at @IREN_Ltd are being promised hefty per diem payments, if they come work "across the road".
Not to say that this is wrong in any way, but they are surely picking up their staff on the fly. But the main issue here, if any, is the seemingly customer concentration that Crusoe/Lancium have with @Oracle, who were also the customer for the site with the gas turbine issue.

Now all these things don't say Crusoe is a bad data center builder, I think they have organized Stargate in a way that is impressive. I have followed it from the first moment this site was chosen, and the speed of construction is admirable.

That said, with speed comes mistakes, and where this "pause" is happening just as land is being graded, I place big question marks around the reasoning of this.

Let's assume in batch zero world, a 40GW pipeline is even remotely accurate, would you really want to walk away from a customer that you have for a pretty significant amount of that available capacity?

My personal believe is that this was another Oracle project, and Oracle just didn't see this happening with Crusoe, after what they delivered in Abilene, and haven't even remotely started to deliver in Childress. (see picture for their "site" to be interconnected in 4 months).

Nothing that's coming from Crusoe should be considered to have any impact on the demand for AI.
I brace myself for Oracle earnings, as there may be a disappointment coming.

The final note from Shanu couldn't be more true:
"Don't underestimate how unpopular AI and datacenters are becoming across the US."

This is unfortunately a reality, but as painful as it is for some, it's pushing up the price for the operators and builders that can deliver.

Long $IREN

[FreeFromMatrix @ViralMuzik1989]

$1.4 Million on 140$ $IREN calls

Smart money has picked the side 👀

[SilentAlert @SilentAlert1]

NEWS 1.8GW Data center paused!

Seem like building critical infrastructure of Data Centers along with 900MW behind the meter and combination of @Bloom_Energy at Wyoming came with abundance of complexity. $BE

This is why $IREN with direct connect and approved power source in given time will have a much more rewarding time to compute.
@FransBakker9812

[Negligible Capital @negligible_cap]

*DATA CENTER DEVELOPER CRUSOE PAUSES AI PROJECT IN WYOMING

Crusoe, a data center developer for hyperscalers, announced they’re pausing development activities of a 1.8GW campus in Wyoming at the request of its customer.

“Crusoe was working with Blackstone Inc.-backed energy company Tallgrass to develop a 1.8-gigawatt campus in Cheyenne, Wyoming, for an undisclosed tenant. “At the request of our customer, Crusoe has paused its development activities” on the site, the company said Tuesday in a statement.””

Customer likely a hyperscaler hitting the pause button. At least partially being attributed to tech dump

[Blockspace @blockspace]

$IREN IS DOWN 13% TODAY 📉

In the last month, the company has announced:

• Plans for 800 MW data center in Australia
• Closing $3.65B GPU financing
• Targets $4.4B ARR at Childress site
• Closed $3B convertible notes
• $3.4B cloud deal with Nvidia

[Mirantis @MirantisIT]

MCP Dev Summit Bengaluru is almost here! 🇮🇳

Shipping an MCP server today still feels pre-Docker. Pick a language, spin up a process, hope the dependencies cooperate.

Bharath Nallapeta, OSPO Team Lead at Mirantis, has a better way.

🎤 "One MCP Server, Five Languages, Zero Containers"
📅 Wednesday 10 June · 1:55-2:20pm IST · Convention Hall

Live on stage, Bharath will build a polyglot MCP server using the WebAssembly Component Model. Then he'll run the same bytes in four environments: a Kubernetes pod, a Spin function, a browser tab, and a Raspberry Pi.

Stay for the honest take on what WASM still can't do. https://t.co/SBQOhnM7sd

[Mirantis @MirantisIT]

Mirantis today announced that PowerX, Inc., a Japanese clean energy company deploying EV charging stations and battery storage systems nationwide, has deployed k0s across hundreds of distributed edge sites in Japan. The deployment establishes a unified Kubernetes control plane spanning from central cloud to edge devices.

PowerX selected k0s to standardize operations across ARM and Intel hardware, replacing fragmented software stacks with a single, lightweight Kubernetes distribution.

Since deploying k0s, PowerX runs leaner and scales faster:
• One platform across heterogeneous device types
• Self-healing infrastructure that keeps sites online
• Open source at the core, with no vendor lock-in

Mirantis continues to invest in compact, production-grade Kubernetes solutions that enable organizations to operationalize distributed infrastructure at scale.

Learn more: https://t.co/vajDxCiFLG

[The God Particle @_Sgr_A_Star]

$IREN

Is IREN trying to corner the Critical Environment Cleaner market? 😂

Jesus!

👀

[mon @moninvestor]

$IREN - Bernstein maintains PT at $100.

As long as the stock stays in this channel, we will hit it.

[The Tech Investor @TheTechInvest]

🚨 BREAKING: CHINA TO INVEST $295B FOR AI BUILDOUT PURPOSES