SemiAnalysis ClusterMAX Lead @JordanNanos explains how Nebius picked up demand CoreWeave couldn’t handle and then executed well enough to reach the Platinum tier of its GPU cloud rankings: “CoreWeave has set the standard. I mean, they are the best in terms of technology.” “These guys deploy 10,000 GPUs a week at the peak when they’re building data centers. That’s incredible.” “But their balance sheet is full. There’s only so many people and resources you can marshal to build out another gigawatt when they’re going to 15 and beyond.” “We have customers coming to us and saying, ‘I saw you write such good things about CoreWeave. I just have $100 million I’m trying to give to them. They’re telling me they can’t accept it until May of next year because they’re backed up.’” “And therefore Nebius has been capitalizing.” “They’ve also executed very well. The technology that they’ve developed is strong.” “We think that managed clusters are approaching this feature-complete world where it’s hard to tell the difference in quality between Nebius and CoreWeave.” “Certainly our testing experience has been really solid on both of them.”

Man i love the bay area. Such a privilege to live here.

Tibo... save me

Now it's on @ReplayRyan

Congratulations to the @CoreWeave team on another exciting NVIDIA Vera Rubin milestone 💚

From an OpenRouter employee, directly.

CoreWeave Engineering might have discovered a ✨problem✨

9:41 is because Steve Jobs showed the first iPhone at that time for the first time - since then it's been kind of a rule. If you have an app on the app store, they also appreciate using 9:41 10:10 shows to a "smile" on watches with hands, and 10:09 is Apple saying they are ahead of the curve very niche info 😉

I found my boy… @adriangrenier @betr

I am pleased to see that Ant’s new Claude Code has a version of background computer use on par with the version from Codex from last May this year. Nice work! Shipping great features first turns out to be a great way to encourage other labs to ship too. We will continue to do this until other labs pay more attention to shipping. This is good for everyone and there is a lot of room left to go! We solved computer use in practice for GPT models about four months ago. But computer use is a significantly awesome thing no matter what model you use, and it is important that the industry similarly spends more effort to train their models to be great at computer use, among other elements of model capabilities. As models become more capable and central to businesses and economies, the value of computer use only increases. We should be taking it seriously, and doing the right thing for our customers and the world.

We are going to need a different AGI benchmark. Where is the goalpost moving next?

Marin 535B-A23B is 13% through training. This hero run would not be possible without the generous support of the Jen-Hsun and Lori Huang Foundation, which provided the funding for the compute (Coreweave). Thanks @JensenHuang for supporting open models!

Never bet against Jensen

A few weeks at OpenAI feel like years at other companies in terms of how much gets done. Crazy how much you can age in a few days

Training shaped the first generation of AI networks. Agentic AI is now rewriting the requirements. @CoreWeave shares five architecture lessons from building an AI network fabric designed to support training, inference, and agentic workloads concurrently at fleet scale. Learn more ⤵️

We only make Flash models now

It was great to have @Raekwon at our office again. Congratulations to the Wu-Tang Clan on their farewell tour. Never forget your OGs.

For those curious, the browser converts non-ASCII domain labels using Punycode, since domain names must be ASCII at the DNS level. So, ▲ = U+25B2, gets Punycode-encoded to xn--c1h So xn--c1h.to resolves to “https://xn--c1h .to” (minus the space)

Whoever invented HEIC format belongs in jail.