Something remarkable happened yesterday as we saw the first signs of life of $AGPU, @AxeCompute. A @NasdaqExchange listed company.
Result: 900x normal volume, with a ยฑ120% share price increase. Off to an impressive start!
A thread on why this is worth investigating.
@atmoio Love the explainer sir, as always
Do you think/expect that quantum computing will have the power to solve the AI-issues you mention in your talks?
And do you think that issues such as 'for every word AI generates it needs to read the whole convo' can be overcome in the future?
When egress is free, you can design infrastructure around performance and resilience.
When it is not, teams quietly constrain their architecture to avoid the next invoice: fewer regions, less replication, more risk.
That trade-off is avoidable with our zero-egress model.
Read more here: https://t.co/YJp8d8taFq
I asked @perplexity_ai to do an analysis. The Verdict?
Use @AethirCloud's Aethir mesh to run your agents. By far the best prices I have seen so far for very strong models.
Aethir Mesh is live.
Top open-source models, including DeepSeek, Kimi, GLM, and more, are available on Aethirโs proprietary LLM API platform.
Plug your Aethir Mesh API key into your Aethir Claw agent and start building.
Cloud GPU or bare metal GPU?
The answer comes down to 4 variables:
1. Workload duration (72-hour training threshold)
2. GPU utilization rate (55-65% crossover point)
3. Data residency requirements
4. MLOps team maturity
Read more: https://t.co/MRHUJcK6X9
$AGPU #AI
Training and inference are two different jobs.
We see enterprise teams running both on the same cluster every week.
That gets expensive fast.
In this article we break down what each job actually demands and where the cost of the wrong mix shows up.
https://t.co/lvU4wMIT6c
Axe Compute (NASDAQ: $AGPU) has received $43M in prepayment against its $260M enterprise GPU contract.
This is the first contracted cash milestone under the agreement and confirms the build is underway and on track for Q3 2026 deployment.
CEO Christopher Miglino - We are committed to continued execution and delivering on schedule, and proving that Axe Computeโs build program does what we said it would do. There are many other prospects for large-scale deployments and we are doing our job to get them launched -
Full release: https://t.co/ff8AadaIlP
Big Tech will spend $725 billion on AI capex in 2026.
McKinsey says the world needs $6.7 trillion in data center investment by 2030 and TSMC is reporting 50% annual growth in AI chip demand.
Everyone is chasing the same constrained supply.
Axe Compute runs 435,000+ GPUs across 93 countries today. Provisioned in 48 hours, at up to 80% below hyperscaler rates.
The capacity exists. The question is who you reserved it with.
Full market analysis: https://t.co/NLGybvNryq
Maybe this reality check finally gets educators to do what they are really passionate about.
Help people become good, skilled and competent young professionals. Instead of assessing boring papers, we can finally focus on actual personal and professional growth.
@atmoio
Good find, I still teach at a Dutch University and wrote a long email about this to my colleagues (will also use your vids, for the AI reality check in a course, if you don't mind).
My view: Education is getting more important than ever, especially in the age of AI.
> AIs are now evaluating AI-generated papers, reducing the entire academic exercise to a conversation between two robots
Wait so now neither the students nor the teachers have to do any work and the students get their diploma with the fancy cursive font and the teachers get paid??
This is the single greatest thing to ever happen to the university industry.
We need to shift our thinking (all of us) to "What kind of professional am I with AI".
And as educators we need to test competencies over products.
Can a student explain how data was analyzed? Can a student elaborate on and defend the presented marketing strategy?
In one month
Strategic compute reserve value up by 117%
$ATH holding up by 121%
$AGPU market cap up by ยฑ170%
And @AxeCompute just announced their $260m 3-year deal.
This is looking good!
A new way to value AI infrastructure is here.
The Axe Compute dashboard gives a real-time view of:
โข Strategic Compute Reserve (SCR)
โข mNAV per share
โข GPU capacity unlocks
All tied to $AGPU
๐ https://t.co/O7OR1HDKSX
#AI#ArtificialIntelligence#Investing#StockMarket #TechStocks #AIInfrastructure #DataCenters #GPU #NASDAQ #AGPU
Because everyone, not just the privileged few, deserves the tools to see the world clearly.
I like economics, and @JustinWolfers is the GOAT - so give @PlatypusEcon a follow :-)
Announcing: @PlatypusEcon โ a Substack, YouTube channel, and podcast built around one idea: Economics is one of the most useful, surprising, and weirdly beautiful subjects in the world.
It belongs to everyone โ not just the people who happened to take it in college. ๐งต
$AGPU in focus today.
CEO Chris Miglino appears on Schwab Networkโs Market on Close following a $260M enterprise GPU deal.
More visibility today, with another appearance on Cheddar later this week.
#AGPU#AI#Stocks#Investing
Axe Compute CEO Chris Miglino on Fox Business breaking down our $260M deal and what it means for the future of AI infrastructure.
Demand for compute isnโt slowing - itโs accelerating. And weโre building for what comes next.
#AGPU#Neocloud#NVIDIA#AIinfrastructure
Watch here โ
$AGPU just closed a $260M enterprise contract โ 2,304 NVIDIA B300 GPUs, dedicated, U.S.-based, deploying Q3 2026.
Enterprise AI buyers are done waiting. They want dedicated infrastructure, their hardware, their location, their terms.
That is exactly what Axe Compute delivers.
$AGPU #AI #neocloud #EnterpriseAI
https://t.co/pMmFxGN3gN