io.net Re-poster

The US has 56GW of data centre capacity.

China has 24GW, with 50GW more in the pipeline.

Most new AI compute is being built in two places.

For everyone else, the question isn't what to build. It's whether you can get the compute.

That's what http://io.net is solving.

Over 20B tokens processed on OpenRouter in a single day.

In July, we were serving around 3B a day.

Same machines. More output.

This is what optimisation and a better tech stack look like.

Built on https://io.net.

Token prices are down ~70% this year.

But bills keep climbing anyway.

The infrastructure behind each model call is a budget decision, not a technical detail.

The fix?

Route work to the right model, choose who runs it, and get more control over costs.

@ionet and @OpenRouter make it possible. Bring your own key, keep your setup.

AI for everyone, not just the biggest budgets.

https://www.forbes.com/sites/rashishrivastava/2026/10/06/ai-costs-too-much-the-reckoning-is-here/

Altman says the world should accept some bad things for the benefits of AI.

Here's another idea.

We wouldn't have to accept them if a few labs and three clouds weren't the only ones running it.

Open compute means fewer trade-offs.

That's what http://io.net is here for.

You have the budget for compute.

But you might still not get the GPUs.

When fleets back billion-dollar loans, lenders favor long contracts over startups paying by the hour.

The financing can decide who gets the chips.

And builders go to the back of the line.

Not with http://io.net.

34M+ compute hours delivered.

Up from 33M in July.

No new data center required. Just GPUs that already existed, put to work.

@ionet lets you use what's built. And pay a fraction of hyperscaler prices while doing it.

62% of Americans back a pause on new AI data centers.

AWS's answer: a 3,000-word blog post.

The better answer: stop pouring concrete.

Millions of GPUs already exist. Many sit idle.

Use what's built before building more.

That's https://io.net.

Your agent writes the code.

Then it tests it, picks a GPU, deploys the container, pays for it, and shuts it down.

Nobody had to file a ticket or get woken up at 2am.

That's what infrastructure built for agents looks like.

@ionet Agent Cloud.
https://io.net/blog/why-ai-agents-need-autonomous-compute-infrastructure

Only 13% of companies are on track with their AI initiatives.

Nearly three-quarters see positive returns.

But fewer than a third get past the pilot.

Pilots are cheap. Production is where the compute bill shows up.

That's the gap https://io.net is built for.

Open weights don't mean open infrastructure.

Run open models on hyperscaler GPUs and you're just swapping one lock-in for another.

It's the same vendor pricing, availability limits, and cost to leave.

The same 70B workload. Four GPUs. Twenty hours a day:
AWS on-demand: around $9,600 a month.
http://io.net: $3,576 a month.

Open models and open compute is what http://io.net was built for.

The White House just renamed AI "Super Intelligence."

New name.

Same GPU shortage.

You can rebrand the model. You can't rebrand the costs or the waitlists.

Whatever you call it, it runs on compute. We're making sure it's available to the many, not the few.

$31.6 trillion

That's what PwC estimates will be spent on AI infrastructure through 2050.

And you still can't get a GPU.

The AI compute crisis isn't just a chip shortage.

It's a coordination problem.

Capacity is reserved, idle, and siloed behind procurement cycles.

More chips alone won't fix that. That's why @ionet is building a different solution.

GPUs don't run on forecasts.

2 GW of US data centre capacity scheduled for 2026 hasn’t even begun construction.

And more than half the GPU servers sold between 2026 and 2028 may lack the power to run.

Hyperscalers are racing to buy more chips while struggling to bring new sites online.

There's another option.

@ionet makes underutilized compute from around the world available now. Not someday.

"There are no adults in this room."

That's a direct quote from an EU AI advisory board member.

Governments are years behind. Labs are writing their own rules.

But the teams building on top carry the risk.

The answer isn't waiting for either to catch up. It's greater participation, access, and transparency.

That's what @ionet was built for.

https://www.nytimes.com/2026/09/27/technology/ai-government-regulation.html

60% lower compute costs.

Same output, one unified API.

That's @kayos_ai, an AI startup running their entire agent stack on io.intelligence.

Previously they were spending $2,500/month per customer across a fragmented setup (Groq, Cerebras, Anthropic, OpenAI).

Now it's $1,000/month, zero rate limits, one provider.

This is what happens when compute stops being the bottleneck.
https://io.net/blog/kayos-case-study

11 billion tokens.

In one day.

That's http://io.net on OpenRouter yesterday. Up from a few billion a day back in July.

Nemotron 3.5 Lightning alone processed 4.82B tokens. GLM 5.3 and GLM 5.3 Flash added another 4.24B combined.

This is what it looks like to have teams routing real workloads through http://io.net, every single day.

$28 million.

Network earnings don't lie.

Real GPU demand, converted into real payouts, every month.

This is what decentralized compute looks like when it's actually working.

Google just put its AI chips in a rocket.

Project Suncatcher is testing whether space can host AI compute at scale.

Read that again.

One of the world's largest hyperscalers is looking past its own data centers because centralized compute, even at Google's scale, has a limit.

The bottleneck was never just chips.

It's who controls that.

Decentralized compute solves this problem today.
Without a rocket.

Every AI conversation hits the same wall.

Compute is scarce, centralized, and controlled by a handful of hyperscalers.

Decentralized compute is the answer.

On this week's Nephology podcast, @ionet's Director of Brand & Comms David Sherman joins host @crypjo091 to make the case.

Centralized compute got us here.

It won't get us to the next stage of AI.
https://www.youtube.com/watch?v=bQ682LGGh8Y

Generative AI hit 53% adoption in three years.

Faster than the PC.

Faster than the internet.

But there's an access problem.

The US hosts 5,427 data centers. More than 10x any other country.

Demand is global.

The supply is concentrated.

http://io.net exists to change that.

Source: Stanford HAI, 2026 AI Index